Diet of key predators responsible for livestock conflict in Namaqualand, South Africa

Transcription

1 Diet of key predators responsible for livestock conflict in Namaqualand, South Africa by Corlé Jansen Thesis presented in partial fulfilment of the requirements for the degree of Master of Science (Conservation Ecology) at the Stellenbosch University Supervisor: Dr. Alison Leslie Co-supervisor: Dr. Quinton Martins Department of Conservation Ecology and Entomology Faculty of AgriSciences March 2016

2 Declaration By submitting this thesis electronically, I declare that the entirety of the work contained therein is my own, original work, that I am the authorship owner thereof (unless to the extent explicitly otherwise stated) and that I have not previously in its entirety or in part submitted it for obtaining any qualification. Date: March 2016 Copyright 2016 Stellenbosch University All rights reserved i

3 Abstract Human-wildlife conflict (HWC) occurs in areas where humans and wildlife occupy the same area or compete for the same resources. Although some carnivores are responsible for incidental attacks on humans, predation on livestock is an increasingly common form of HWC. Understanding the ecology of these predators outside the confines of protected areas could provide insight into decreasing conflict and ensuring the persistence of these animals in non-protected areas. I analysed the diet of leopard (Panthera pardus), caracal (Caracal caracal) and black-backed jackal (Canis mesomelas) in Namaqualand, Northern Cape, South Africa, an area where HWC is commonly reported. Scats were collected for each predator in both protected areas (Namaqua National Park) and surrounding, nonprotected farmlands (810 km²). Eight caracals were also collared to analyse caracal diet from GPScluster data. The diet of the three predators was assessed across both land classifications (protected vs. non-protected). Prey availability was determined by means of camera and small mammal traps and compared between the two land classifications. The relative abundance index (RAI) was used to determine the abundance of prey species on the two contrasting land classifications and whether prey abundance and availability influenced the feeding ecology of leopard, caracal and jackal in Namaqualand. All three predators relied on the most abundant and easy to catch prey species, reflecting opportunistic feeding behaviour. In the protected national park, where livestock was absent, all three predators selected for natural prey items. These findings coincided with previous studies on leopard, caracal and black-backed jackal in South Africa. A shift in leopard diet was observed on farmlands, as livestock replaced small-to medium-sized ungulates in scats. For blackbacked jackals, steenbok (Raphicerus campestris) contributed >20% to the total biomass consumed in protected areas while on farmlands sheep (Ovis aries) contributed > 20% and steenbok only < 5%. These findings in scat are mirrored in ungulate surveys; steenbok was the most abundant small-to medium-sized ungulate in the national park and sheep were the most abundant prey on farmlands. Caracal preferred hyrax (Procavia capensis) and lagomorpha as prey, while predation on livestock occurred in low frequencies (scat analysis, 6.9%), making caracal the predator which depended the least on livestock. Land-use also had very little effect on caracal diet. When analysing caracal diet by means of kill site analysis, sheep contributed the bulk to the total biomass consumed (59.5%). However, GPS cluster analysis is inherently biased towards the overestimation of larger bodied prey items and excludes smaller prey items (< 1 kg) which contributed > 25% to the total biomass consumed according to scat analysis. Predation of livestock by these three predators was not significant in relation to livestock availability on farmlands, especially for caracal. Due to the ii

4 opportunistic feeding behaviour of these predators it was more likely that livestock was an alternative prey source. A suitable natural prey base on farmlands would decrease livestock losses, especially where leopards depredate on stock. Leopards are the last remaining large predator in this area and the loss of these large felids could be detrimental to the healthy functioning of the ecosystem. If increased vigilance is practiced during the lambing period, lambs could survive to past their vulnerable size when they fall victim to jackals. Improved livestock husbandry methods, implementation of guarding animals and herders and various other holistic methods could decrease livestock losses in Namaqualand. iii

5 Opsomming Konflik tussen mense en diere kom gewoonlik voor in areas waar wilde diere en mense dieselfde area beset en vir soortgelyke hulpbronne kompeteer. Deur die ekologie van roofdiere kwesbaar vir hierdie konflik buite beskermde gebiede te bestudeer, kan baie insigte verkry word oor hoe om konflik te vermy en te verminder. Verder kan dit verseker dat sulke predatore nie verlore gaan buite formeel beskermde areas nie. Hierdie studie het die dieet van luiperd, rooikat en rooijakkals in Namakwaland, Noord-Kaap, Suid-Afrika geanaliseer. Mismonsters van die drie predatore is versamel in die Namakwa Nasionale Park en omliggende veeplase (810 km²). Agt rooikatte was ook voorsien met radio-nekbande om rooikat dieët verder te ontleed. Die dieet van die drie predatore is geanaliseer en vergelyk tussen beide die nasionale park en die omliggende plase. Die beskibaarheid van prooi op altwee grondgebruike is ook ontleed deur gebruik te maak van kameras in die veld en klein-soogdier lokvalle. n Relatiewe volopheids-indeks (RVI) is gebruik om te bepaal of die beskibaarheid en getalle van prooi die dieet van luiperd, rooikat en rooijakkals beïnvloed in Namakwaland. Al drie van die predatore het opportunistiese voergedrag getoon. Die dieet was verder grootliks afhanklik van die volopheid van maklik-bekombare prooi. Vorige studies van luiperd-, rooikat- en rooijakkals-dieet stem ooreen met die resultate verkry van die mismonsters versamel in die nasionale park. n Verskuiwing in luiperddieet is op kleinveeplase waargeneem waar vee essensieël die rol van kleiner bok-soorte vervang. n Soortgelyke tendens is waargeneem in jakkalsdieet. Steenbok, wat > 20% bygedra het tot die algehele biomassa gevreet deur jakkals in die nasionale park, is effektief vervang deur skaap, wat > 20% bygedra het tot biomassa op aanliggende plase, waar steenbok slegs > 4%. Steenbok was ook die volopste van die kleiner bok-soorte in die nasionale park, met skaap die volopste prooi item op die plase. Rooikat het dassie en lagomorpha (hase en konyne) verkies in beide die nasionale park en die aanliggende plase. Rooikat het selde gevoed op vee; op 6.9% die laagste van al die predatore in die studie. Rooikat-dieet is ook ontleed van karkasse wat opgespoor is deur middel van GPS-kluster besoeke. Die data verkry bewys dat skaap n beduidende deel van die algehele biomasssa gevreet deur rooikat bydra. Hierdie metode het egter alle prooi kleiner as 1 kg, wat meer as 25% bygedra het tot die algehele biomassa gevreet deur rooikat, uitgesluit. Luiperd was die hoofpredatoor van boerbokke in die droeë seisoen. Volgens die beskikbaarheid van vee was die predasie deur hierdie drie predatore op vee nie beduidend nie. Omdat hierdie predatore so n aanpasbare patroon volg, in terme van hul dieet, is dit meer waarskynlik dat luiperd, jakkals en meer so rooikat vee gevang het as alternatiewe prooi. As daar op iv

6 plase n voldoende, natuurlike prooi-basis beskikbaar is sal vee verlieste moontlik verminder, veral verlieste as gevolg van luiperd predasie. Luiperds is die laaste oorblywende groot karnivore in hierdie area en die moontlike verlies van hierdie diere sal groot nagevolge hê vir die algehele funksionering van Namakwaland as n gesonde, interaktiewe ekosisteem. As daar meer waaksaamheid uitgeoefen word in tye wanneer ooie lam, kan lammers grootliks ongehinderd groei tot op n punt waar jakkalse hulle nie meer sal teiken nie. Uiteindelik kan dit lei tot n wen-wen situasie vir beide die ekosisteem se gesonde funksionering en produksie. n Toename in vee-bestuur metodes, soos die gebruik van herders in die veld, vee waghonde en ander holistiese metodes kan vee verlieste in Namakwaland effektief verminder. v

7 Acknowledgements Working on this project has been a privilege. In these past two years I was fortunate enough to contribute to a cause which I believe in. I would like to express my gratitude to everyone involved in making this project a reality: To my supervisor, Dr Alison Leslie, for all the guidance and support throughout my research. My co-supervisor, Dr Quinton Martins who without this project would never have happened. His passion for cape leopard conservation was the driver behind the PEACE project. Dr Bogdan Cristescu and Kristine Teichman, the parents. They provided me with constant advice and support throughout my time in the field. Without their guidance and suggestions I would have been lost. I can t thank you enough. The Cape Leopard Trust for providing logistical support, financial support and most importantly, a platform to do what we love. Woolworths for the biggest funding the project had and willingness to help protect our biodiversity and support local farmers. Bryan Havemann and Helen Turnbull from CLT for their constant logistical support and dealing with endless requests. Bernard and Elanza van Lente and the SANParks team who not only provided us with housing, but also constantly came to our aid when windpumps needed to be fixed and bakkies to be nudged off boulders. Conservation South Africa for being part of the solution. The private landowners Nathan Aggenbag, Japie van Wyk, Gert van Wyk, Marius van Wyk, Flippie Kennedy, Corrie van der Westhuizen en Phillip, Pieter, Ivan van Niekerk, Robbie en JJ Archer, die Strydoms en almal anders. Ek weet dit is nie altyd maklik om saam met die groenes te werk nie, maar ons waardeer al die tyd, hoop en geduld wat julle in hierdie projek ingesit het. Ons hoop dit wys positiewe resultate. vi

8 All of our volunteers; the Americans Chris Vlautin (sorry, Dr Vlautin), Ryan Wilbur, Alex Potash, Bradley (Bo) Larson, Jessica Gomez, Sam Childs, Kenneth Loonam, Jared Franklin and Natasha Bowman; the English ladies Charlotte Beaton and Carrie Dunford; the Belgian Sam Puls and our one South African, Elbé Visser. You guys have been amazing! I couldn t have done this without you (thanks for keeping me sane). All our volunteer vets (who doubled as research technicians), especially Victor Lizana Martin and Luca Mendes. Anita Meyer and Jeannie Hayward from CLT. Anita helped with all things scat and without her help I would not have known what to do. Die locals van Kamieskroon en Springbok wat vir ons en al die uitlandse vrywillegers gewys het wat gasvryheid is. Agrimark Springbok for always helping me when I pitch up with a weird list of things to buy; Supaquick Springbok and Autorama Springbok for all the tyres fixed. The Nematology department and Dr Malan and her team for allowing me to work on the microscope. Dr Dan Parker and Rhodes University for providing images to build a reference collection Sasha-Lee Maggs for spending her free time in the lab, washing and sorting scat. You helped me so much to get ahead of time. Rudi Swart for always being there when I needed him and for dealing with all the mood swings that came with thesis writing. And finally my amazing family, especially my dad. Sonder julle sou dit glad nie moontlik gewees het om hierdie projek te doen nie. Julle ondersteuning finansieel en emosioneel is iets wat ek nooit sal vergeet of prys gee nie. Pappa, jammer oor die bakkie maar die Triton het weereens homself bewys! vii

9 Table of Contents Declaration... i Abstract... ii Opsomming... iv Acknowledgements... vi Table of Contents... viii List of Figures... xiii List of Tables... xvi List of Appendices...xix Chapter 1: Literature Review and Study Area Human-wildlife conflict (HWC): a worldwide problem Human-carnivore conflict (HCC) Livestock predation: The problem Human-carnivore conflict in South Africa: with a focus on livestock losses Importance of diet in carnivore studies and mitigating HCC Focal Species The leopard (Panthera pardus) an apex predator The mesocarnivores The caracal (Caracal caracal) The black-backed jackal (Canis mesomelas) Study Area Location and History Climate Geology and Soils Vegetation viii

10 1.6 Objectives of this study References Chapter 2: Methodology Data Collection Diet Estimation through Scat Collection and Analysis Caracal capture and immobilisation Diet Estimation through GPS Radio-collar Cluster Visitation Prey Abundance Estimation through Camera Trapping Prey Abundance Estimation through Small Mammal Trapping Data Analysis Diet Estimation through Scat Analysis Diet Estimation through GPS Radio-collar Cluster Visitation Prey Abundance and Preference Analysis Statistical Analysis Diet Statistical Analysis Scat and GPS Radio-collar Cluster Visitation Statistical Analysis Prey abundance and Preference Statistical Analysis References Appendices Chapter 3: The diet of leopard (Panthera pardus) in Namaqualand, South Africa Abstract Introduction Aims and Objectives Methods Study Area Data Collection Data Analysis ix

11 Scat Analysis Prey Abundance and Preference Analysis Statistical Analysis Results Leopard diet Namaqua National Park versus surrounding farms Prey abundance and preference Discussion General diet of leopards in Namaqualand Namaqua National Park versus surrounding farmlands Conclusion References Appendices Chapter 4: The diet of caracal (Caracal caracal) in Namaqualand, South Africa Abstract Introduction Aims and Objectives Methods Study Area Data Collection Data Analysis Scat Analysis Prey Abundance and Preference Analysis Statistical Analysis Results Caracal Diet Namaqua National Park versus surrounding farmlands x

12 Prey abundance and preference Discussion General diet of caracal in Namaqualand Namaqua National Park diet versus surrounding farmlands Conclusion References Appendices Chapter 5: The diet of black-backed jackal (Canis mesomelas) in Namaqualand, South Africa Abstract Introduction Aims and Objectives Methods Study Area Data Collection Data Analysis Scat Analysis Prey Abundance and Preference Analysis Statistical Analysis Results Black-backed jackal Diet Namaqua National Park versus surrounding farmlands Prey abundance and preference Discussion General diet of black-backed jackal Namaqualand Namaqua National Park versus surrounding farmlands Conclusion References xi

13 5.8. Appendices Chapter 6: Caracal (Caracal caracal) diet: scat analysis or GPS cluster visitation? Abstract Introduction Aims and Objectives Methods Study Area Data Collection Data Analysis Diet Estimation through Scat Analysis Diet Estimation through GPS Radio-collar Cluster Visitation Prey Preference Analysis Statistical Analysis Results Diet Estimation through Scat Analysis Diet Estimation through GPS Radio-collar Cluster Visitation GPS Radio-collar Cluster Visitation versus Scat Analysis Prey Preference Discussion Conclusion References Appendices Chapter 7: Integrated discussion of results and proposed management implications Predator diet in Namaqualand - what have we learnt from this study? Possible solutions to decrease livestock depredation In Namaqualand References xii

14 List of Figures Figure 1.1. Figure 1.2. Figure 1.3. Figure 1.4. Figure 1.4. Figure 1.6. Figure 1.7. Figure 1.8. Figure 2.1. Suitable leopard habitat still available in South Africa according to Swanepoel et al. (2012) developed from a model with various environmental variables, a) excluding human variables, b) estimated human impact and c) where habitat suitability index represents logistic probabilities of occurrences. Negative values indicate areas where human impact has a negative effect on leopard habitat, positive values where human impact had a positive effect and zero values where leopard habitat suitability was not influence by human impacts. A map of South Africa (insert) showing the location of the study area (marked as the grid). Average monthly rainfall for the study area over a 5 year period from (data from Skilpad in Namaqua National Park). Average monthly maximum and minimum temperature for the study area over a 7 year period from (data from Skilpad in Namaqua National Park). The landscape of Namaqualand, consisting of medium to large granite gneiss and flatter valleys. Corlé Jansen The large rock domes of granite gneiss which can be seen in the study area. Corlé Jansen The fields in front of Skilpad, Namaqua National Park in the dry months (December- May). Corlé Jansen The fields in front of Skilpad, Namaqua National Park in the wet months (June- November). Corlé Jansen Map illustrating all locations in study area where leopard (yellow), caracal (red) and black-backed jackal (blue) scat was collected Figure 2.2. Map illustrating all locations in study area camera trapping occurred each cell represents 9 km² with two camera stations for each grid cell. 50 Figure 3.1 Figure 3.2. Prey classes recorded in leopard scat (n=82) collected in Namaqua National Park (n=28) and on surrounding farmlands (n=54), Northern Cape, South Africa. CFO (%) was calculated as the number of occurrences per scat divided by the total number of scats collected. Prey relative abundance index (RAI) calculated from camera trap data collected from March 2014 April RAI was calculated as the total xiii

15 Figure 3.3. Figure 3.4. Figure 4.1 Figure 4.2. Figure 4.3. Figure 5.1. Figure 6.1 detections of a certain mammalian species, multiplying by 100 (to calculate the number of photo captures per 100 trap nights), and dividing by the total number of trap nights. Species accumulation curve (100 randomised iterations) for the entire study area (ICE Mean = 29; ACE Mean = 29), in Namaqua National Park (ICE Mean = 27.8; ACE Mean = 27.4) and the surrounding farmlands (ICE Mean = 29; ICE Mean = 29) of the 29 wild mammal prey items 1 kg in weight and livestock in the study area. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species. A D-value close to 0 indicate prey consumption in proportion to prey availability, (prey items was neither preferred, nor avoided).the biomass consumed and the corrected frequency of occurrence (%) used to calculate the D-value are illustrated. Prey relative abundance index (RAI) calculated from small mammal trapping data collected from September 2014 November RAI was calculated as the total number of detections of a small mammal species, multiplied by 100 (to calculate the number of individuals captured per 100 trap nights), and divided by the total number of trap nights. Insectivores included Elephantulus spp., Macroscelides proboscideus and Soricidae. Species accumulation curve (100 randomised iterations) for small mammal trapping in the entire study area (ICE Mean = 9.37; ACE Mean = 10.24), in Namaqua National Park (ICE Mean = 12.47; ACE Mean = 14.93) and the surrounding farmlands (ICE Mean = 14.33; ICE Mean = 14.24) of the 8 small mammal species trapped in the study area. Sampling efficiency was recorded for all sampling occasions which included 16 traps per site trapping for 3 nights. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species in Namaqualand, South Africa. Both the biomass and the corrected frequency of occurrence (%) used to calculate the D-value is illustrated. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species in Namaqualand, South Africa. Both the biomass and the corrected frequency of occurrence (%) were used to calculate the D-value for black-backed jackal prey preference. Please see Appendix 4M for a table comparing the both CFO and biomass as units used in Jacobs index calculations. Corrected frequency of occurrence (CFO) % for all prey items identified from the 250 scats analysed and 91 kill sites visited in Namaqua National Park and surrounding farmlands, Namaqualand, Northern Cape, South xiv

16 Africa. Figure 6.2. Comparison of scat collection methods showing the frequency of occurrence (FO) of all prey classes identified from the 250 scats analysed from Namaqua National Park and surrounding farmlands, Namaqualand, Northern Cape, South Africa. Figure 6.3. Relative biomass consumed (%) for all prey items identified from the 250 scats analysed and 91 kill sites visited in Namaqua National Park and surrounding farmlands, Namaqualand, Northern Cape, South Africa. The relative biomass consumed for scats analysed is the equivalent of corrected biomass consumed for kill site analysis. 224 Figure 6.3. Figure 6.4. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species identified from caracal kill sites in Namaqua National Park and the surrounding farmlands, Namaqualand, Northern Cape, South Africa. D-values are based on corrected frequency of occurrence (CFO) % of prey items from caracal scat and caracal kill sites. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species identified from caracal kill sites in Namaqua National Park and the surrounding farmlands, Namaqualand, Northern Cape, South Africa. D-values are based on the relative and corrected biomass consumed (%) of prey items from caracal scat and caracal kill sites xv

17 List of Tables Table 3.1. Table 3.2. Table 3.3. Table 3.4. Table 3.5. Table 3.6. Table 4.1. Prey classes prey species recorded in leopard scat (n=82) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Frequency of occurrence (FO) (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=100). Corrected frequency of occurrence (CFO) (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=82). For a table with all species identified see Appendix 2A. Biomass consumed calculated from leopard scat (n=82) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 2B. Prey classes and prey species recorded in leopard scat collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. FO (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences. CFO (%) was calculated as the number of occurrences per scat divided by the total number of scats collected. For a table containing a full list of species identified see Appendix 2C (Namaqua National Park) and Appendix 2D (farmlands). Biomass consumed calculated from leopard scat (n=28) collected in Namaqua National Park, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 2E. Biomass consumed calculated from leopard scat (n=54) collected on farmlands in Namaqualand, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 2F. Relative abundance index (RAI) of all mammalian species recorded on the camera traps in both Namaqua National Park and the surrounding farmlands in Namaqualand, Northern Cape. The corrected frequency of occurrence (CFO) used in Jacobs Index calculations for each separate land-use is also summarised. See Appendix 2K (CFO) and Appendix 2L (biomass consumed) for comparative figure of D-values calculated for the national park and farmlands. Prey items recorded in caracal scat (n=250) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Frequency of occurrence (FO) (%) was calculated as the number of occurrences of each prey xvi

18 Table 4.2. Table 4.3. Table 4.4. Table 4.5. Table 4.6. Table 5.1. Table 5.2. item divided by the total number of occurrences (n=327). Corrected frequency of occurrence (CFO) (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=250). For a table with all species identified see Appendix 3A. Biomass consumed calculated from caracal scat (n=250) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 3B. Prey items recorded in caracal scat collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. FO (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences. CFO (%) was calculated as the number of occurrences per scat divided by the total number of scats collected. For a table containing a full list of species identified see Appendix 3C (Namaqua National Park) and Appendix 3D (farmlands). Biomass consumed calculated from caracal scat (n=98) collected in Namaqua National Park, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 3E. Biomass consumed calculated from caracal scat (n=152) collected on farmlands in Namaqualand, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 3F. Relative abundance index (RAI) of all mammalian species recorded on the camera traps in both Namaqua National Park and the surrounding farmlands in Namaqualand, Northern Cape. The corrected frequency of occurrence (CFO) used in Jacobs Index calculations for each separate land-use is also summarised. See Appendix 3K (CFO) and Appendix 3L (biomass consumed) for comparative figure of D-values calculated for the national park and farmlands. Prey items recorded in black-backed jackal scat (n=196) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Frequency of occurrence (FO) (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=336). Corrected frequency of occurrence (CFO) (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=196). For a table with all species identified see Appendix 4A. Biomass consumed calculated from black-backed jackal scat (n=196) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 4B xvii

19 Table 5.3. Table 5.4. Table 5.5. Table 5.6. Table 6.1. Table 6.2. Table 6.3. Table 6.4. Prey items recorded in black-backed jackal scat collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. FO (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences. CFO (%) was calculated as the number of occurrences per scat divided by the total number of scats collected. For a table containing a full list of species identified see Appendix 4C (Namaqua National Park) and Appendix 4D (farmlands). Biomass consumed calculated from black-backed jackal scat (n=94) collected in Namaqua National Park, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 4E. Biomass consumed calculated from black-backed jackal scat (n=102) collected on farmlands in Namaqualand, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 4F. Relative abundance index (RAI) of all mammalian species recorded on the camera traps and captured with small mammal trapping in both Namaqua National Park and on the surrounding farmlands in Namaqualand, Northern Cape. The corrected frequency of occurrence (CFO) used in Jacobs index calculation for black-backed jackal on each separate land-use is also shown. For figures illustrating the difference in Jacobs index compared between the two land-uses please see Appendix 4K (CFO) and Appendix 4L (biomass). Prey items recorded from GPS cluster visitations in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. FO (%) was calculated as the number of occurrences of the respective prey item across kill sites divided by the total number of kill sites identified (n=91). Biomass consumed calculated from caracal kill sites (n=82)* visited in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Both the biomass consumed and the corrected biomass consumed is presented. See footnote for more details regarding calculations. Prey classes recorded in caracal scat collected (n=250) in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa and kill sites visited (n=91). CFO (%) was calculated as the number of occurrences per scat or kill site divided by the total number of scats collected or kill sites visited. CFO for scat analysis is similar to the FO of kill sites. Biomass consumed calculated from caracal scat (n=250) collected and kill sites visited (n=91) in Namaqua National Park and on farmlands in Namaqualand, Northern Cape, South Africa. For a Table with all species and calculations based on scats analysed please see Appendix 3C, whereas for calculation based on kill sites see Table xviii

20 List of Appendices Appendix 2A Appendix 2B Appendix 2C Datasheet filled out at each caracal capture, as provided by Dr Quinton Martins. Although the original sheet was set-up for leopard captures, the same rules apply to caracal capture. GPS cluster visitation searching method on a 50 m radius. A zigzag search pattern was used, with each person starting at the centroid and walking out to the edge of the 50 m radius. Photographic example of hyrax (Procavia capensis) kill remains (left) and the data sheet filled out for each kill site visited (right) Appendix 2D Small mammal trapping outline. 67 Appendix 2E Appendix 3A Appendix 3B Appendix 3C Appendix 3D Sex, weight and estimated age at time of capture for the 8 caracal radiocollared for this study. Prey items recorded in leopard scat (n=82) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Frequency of occurrence (FO) (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=100). Corrected frequency of occurrence (CFO) (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=82). Biomass consumed calculated from leopard scat (n=82) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Prey items recorded in leopard scat collected in Namaqua National Park, Northern Cape, South Africa. Frequency of occurrence (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=39). Corrected frequency of occurrence (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=28). Prey items recorded in leopard scat collected on farmlands in Namaqualand, Northern Cape, South Africa. Frequency of occurrence (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=61). Corrected frequency of occurrence (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=54) xix

21 Appendix 3E Appendix 3F Appendix 3G Appendix 3H Appendix 3I Appendix 4A Appendix 4B Appendix 4C Appendix 4D Biomass consumed calculated from leopard scat (n=28) collected in Namaqua National Park, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Biomass consumed calculated from leopard scat (n=54) collected on farmlands in Namaqualand, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species in Namaqua National Park and the surrounding farmlands. D-values are based on corrected frequency of occurrence (%) of prey items from leopard scat. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species in Namaqua National Park and the surrounding farmlands. D-values are based on the total biomass consumed of prey items from leopard scat. Jacobs index (D-value) of all prey items found in leopard scats collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. The corrected frequency of occurrence (CFO) % used in Jacobs index calculation and the biomass consumed are included, as well as the Relative Abundance Index (RAI). Prey items recorded in caracal scat (n=250) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Frequency of occurrence (FO) (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=327). Corrected frequency of occurrence (CFO) (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=250). Biomass consumed calculated from caracal scat (n=250) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Prey items recorded in caracal scat collected in Namaqua National Park, Northern Cape, South Africa. Frequency of occurrence (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=130). Corrected frequency of occurrence (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=98). Prey items recorded in caracal scat collected on farmlands in Namaqualand, Northern Cape, South Africa. Frequency of occurrence (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=195). Corrected frequency of occurrence (%) was calculated as the number of occurrences per scat divided by the total number of scats xx

22 Appendix 4E Appendix 4F Appendix 4G Appendix 4H Appendix 4I Appendix 5A Appendix 5B Appendix 5C Appendix 5D collected (n=152). Biomass consumed calculated from caracal scat (n=98) collected in Namaqua National Park, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Biomass consumed calculated from caracal scat (n=152) collected on farmlands in Namaqualand, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species in Namaqua National Park and the surrounding farmlands. D-values are based on corrected frequency of occurrence (%) of prey items from caracal scat. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species in Namaqua National Park and the surrounding farmlands. D-values are based on the total biomass consumed of prey items from caracal scat. Jacobs index (D-value) of all prey items found in caracal scats collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. The corrected frequency of occurrence (CFO) % used in Jacobs index calculation and the biomass consumed are included, as well as the Relative Abundance Index (RAI). Prey items recorded in black-backed jackal scat (n=196) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Frequency of occurrence (FO) (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=336). Corrected frequency of occurrence (CFO) (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=196). Biomass consumed calculated from black-backed jackal scat (n=196) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Prey items recorded in black-backed jackal scat collected in Namaqua National Park, Northern Cape, South Africa. Frequency of occurrence (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=182). Corrected frequency of occurrence (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=94). Prey items recorded in black-backed jackal scat collected on farmlands in Namaqualand, Northern Cape, South Africa. Frequency of occurrence (%) was calculated as the number of occurrences of each prey item divided by xxi

23 Appendix 5E Appendix 5F Appendix 5G Appendix 5H Appendix 5I Appendix 6A the total number of occurrences (n=156). Corrected frequency of occurrence (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=102). Biomass consumed calculated from black-backed jackal scat (n=94) collected in Namaqua National Park, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Biomass consumed calculated from black-backed jackal scat (n=102) collected on farmlands in Namaqualand, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species in Namaqua National Park and the surrounding farmlands. D-values are based on corrected frequency of occurrence (%) of prey items from black-backed jackal scat. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species in Namaqua National Park and the surrounding farmlands. D-values are based on the total biomass consumed of prey items from black-backed jackal scat. Jacobs index (D-value) of all prey items found in black-backed jackal scats collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. The corrected frequency of occurrence (CFO) % used in Jacobs index calculation and the biomass consumed are included, as well as the Relative Abundance Index (RAI). Jacobs index (D-value) of all prey items found at caracal kill sites in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. The corrected frequency of occurrence (CFO) % used in Jacobs index calculation and the biomass consumed are included, as well as the relative abundance index (RAI) xxii

24 Chapter 1: Literature Review and Study Area Chapter 1: Literature Review and Study Area 1.1. Human-wildlife conflict (HWC): a worldwide problem Human wildlife conflict (HWC) can be loosely defined as an interaction where humans and wildlife occupy the same area and/or compete for the same resources (Inskip and Zimmerman 2009; Li, Buzzard, Chen and Jiang 2013). In recent times it is not uncommon for humans and wildlife to come into conflict with one another (Treves and Karanth 2003; Li et al. 2013; Campbell et al. 2014). The human population is growing at an exponential rate which results in an increase in agricultural activities, urbanisation, increased disease transmission between wild- and domestic animals and a further decrease in food resources for wildlife and diminishing natural habitat (Pettigrew et al. 2012). Wildlife is either forced out of their historical ranges or continue to live in close proximity to humans (Treves and Karanth 2003; Pettigrew et al. 2012; Kiffner et al. 2014). There is a general misconception that this conflict is restricted to poverty-stricken areas, however HWC is a worldwide problem (Madden and McQuinn 2014). HWC often results in negative impacts towards either humans, wildlife or both parties (Dickman, Macdonald and Macdonald 2011; Pettigrew et al. 2012). While human-wildlife conflict has become synonymous with carnivores (Dickman et al. 2011), there are various other examples of HWC worldwide ranging across various animal species. For example, elephants in Africa are notorious problem animals due to the damaging of crops, destruction of artificial water sources such as water tanks and the raiding of food stores (Hoare 1999; Parker and Osborn 2006; Taruvinga and Mushunje 2014). Human-elephant conflict has resulted in retaliatory killings of elephants, however great conservations efforts have been made to mitigate humanelephant conflict in both Africa and India (Parker and Osborn 2006; Jadhav and Barua 2012; Mariki, Svarstad and Benjaminsen 2015). Baboons (Papio ursinus) in the Cape Peninsula, South Africa, are also responsible for property damage and the harassment of people in order to obtain alternate food sources (Hoffman and O Riain 2012). In North America, grizzly bears (Ursus arctos) and black bears (Ursus americanus) have adapted to the increasing human population and have learnt to gain from anthropogenic food items that are easily accessible (Wilson et al. 2005; Don Carlos, Bright, Teel and Vaske 2009). In Africa, many species are responsible for fatal attacks on humans (Lamarque et al. 2009). Crocodiles (Crocodylus niloticus) and hippopotamuses (Hippopotamus amphibius) are the large animals responsible for the most fatal attacks on humans in Africa. This is a much less common form of HWC in Africa than crop damage by mammals, but it is still of great concern to many local people living in close proximity to these animals (Lamarque et al. 2009). Internationally snakes are 1

25 Chapter 1: Literature Review and Study Area perceived to be dangerous which leads to either accidental or intentional snake mortality (Bonnet, Naulleau and Shine 1999; Brown, Bishop and Brooks 2009). Free-roaming wildlife is also responsible for a large number of vehicle collisions in which wildlife and sometimes humans perish (Mouron; Morelle, Lehaire and Lejeune 2013; 2014). Disease transmission from wild to domestic animals is also of great concern and another factor responsible for HWC (Bengis, Kock and Fisher 2002; Wilkinson, Smith, Delahay and Cheeseman 2004). In the United Kingdom, culls were implemented to control badger (Meles meles) populations as these animals contribute to the spread of bovine tuberculosis (Donnelly et al. 2006; Byrne et al. 2014). A seemingly unlikely animal species that has also been involved in HWC is the beaver (Castor fiber), mostly due to the misconception that these animals have a negative effect on the environment (Czech and Lisle 2003) Human-carnivore conflict (HCC) Human-carnivore conflict (HCC) has become one of the most urgent conservation problems and is reportedly increasing in frequency (Jackson and Wangchuk 2004; Inskip and Zimmerman 2009; Pettigrew et al. 2012; Johansson et al. 2015). Carnivores play an important role in ecosystem functioning and exert top-down effects influencing other aspects of an ecosystem (Estes et al. 2011). However, the rapid growth and development of the human population is threating the persistence of these free-ranging species across the world (Dar, Minhas, Zaman and Linkie 2009; Inskip and Zimmerman 2009). In many areas carnivores are living in close proximity to human populations (Inskip and Zimmerman 2009). In general, a negative attitude in humans towards carnivores still exists, however the aesthetic value and possible economic benefit of carnivores is being increasingly recognised by communities (Campbell et al. 2014). HCC is mutually detrimental to both humans and carnivore communities. In various regions throughout the world carnivores can threaten human lives through aggressive behaviour and sometimes fatal attacks (Loveridge, Wang, Frank and Seidensticker 2010). In the 1890 s, two lions from the Tsavo region, Kenya received infamy for the killing of between 28 and 135 people and became known as the man-eaters of Tsavo (Patterson 1907; Yeakel et al. 2009). In other parts of the world tigers (Panthera tigris) have been labelled as the felid responsible for the most human deaths (Siddiqi and Choudhury 1987; Karanth and Gopal 2005). In the 1920 s, 7000 deaths were recorded over a five-year period as a result of fatal tiger attacks in India (Peterhans and Gnoske 2001). In more recent times tiger attacks on humans have become less frequent; however livestock depredation is still a common occurrence (Johnson, Vongkhamheng, Hedemark and Saithongdam 2006). The livelihood of local people can also be threatened by carnivores as these animals tend to predate on 2

26 Chapter 1: Literature Review and Study Area livestock when living in close proximity to humans and agricultural land (Balme, Slowtow and Hunter 2009; Thorn, Green, Scott and Marnewick 2013). Wang (2008) reported that tigers and leopards (Panthera pardus) in Bhutan were responsible for losses of 17% of the annual income of households living around Jigme Sinyq Wangchuck National Park. Lions (Panthera leo) living close to Tsavo National Park in Kenya were responsible for the loss of 433 heads of livestock over a four-year period (Patterson, Kasiki, Selempo and Kays 2004; Lamarque et al. 2009). The depredation of livestock by carnivores leads to lethal persecution of many wild carnivores. The retaliatory killing of carnivores has become a common occurrence in areas where people live close to or sometimes within protected areas (Woodroffe, Thirgood and Rabinowitz 2005). Snow leopards (Panthera uncia) in the Tost Mountains of Mongolia were also found to predate on livestock, but this only made up 27% of the total diet (Johansson et al. 2015). Snow leopards occur in areas with rugged terrain and a low abundance of wild prey populations. With the increase in human activities in these rugged areas, livestock populations are becoming more abundant providing an alternate food source for snow leopards (Mishra et al. 2003; Jackson, Mishra and McCarthy 2010). The lethal persecution of these animals has become problematic due to their endangered status, prompting various conservation efforts to decrease livestock depredation and stabilise human-wildlife conflict in these areas (Jackson et al. 2010). Carnivores have been persecuted for predating not only on livestock, but also game species where game ranching is a source of income (Inskip and Zimmerman 2009; Loveridge et al. 2010). Bears are known to be the cause of fatal human attacks, as well as contributing to livestock depredation. In India, the Asiatic black (Ursus thibetanus) and brown bears (Ursus arctos) were responsible for 3 human casualties and the killing of 355 head of livestock in the Great Himalayan National Park Conservation Area from 1989 to 1998 (Chauhan 2003). Sloth bears (Melursus ursinus), occurring in the forests of India, are known to attack humans. These bears rarely feed on humans, but are mostly responsible for attacking and mauling people when coming into contact with people (Rajpurohit and Krausman 2000). Bears also prey on livestock in North America, along with wolves (Canis lupus) and coyotes (Canis latrans) [Wagner, Schmidt and Conover 1997]. There have been incidences where wolves also attack people, either due to being infected with the rabies virus, as a defensive mechanism or sometimes even predatory. In the latter case it is mostly older, single wolves or sometimes certain packs that exploit humans as a food source (Linnell et al. 2002). Leopards are also known for fatal attacks on humans in both Africa and Asia (Loveridge et al. 2010). In India from the Panar leopard killed 400 people an example of just one of the infamous leopards in the world responsible for human deaths (Peterhans and Gnoske 2001). Attacks 3

27 Chapter 1: Literature Review and Study Area on humans by leopards have decreased however in some areas it still occurs. Leopards, being the smallest of the large felids (> 50 kg), tend to attack woman and children, mostly avoiding men (Peterhans and Gnoske 2001). Leopards also contribute to livestock losses, mostly preying on small stock such as sheep and goats or the young of larger livestock such as cattle. In a two-year period ( ) in the Kweneng district of Botswana, 857 leopard attacks on livestock were recorded (Schiess-Meier, Ramsauer, Gabanapelo and Köning 2007). In Kenya from leopards killed on average 4.3 cattle and 10.5 sheep per year (Mizutani 1999). Jaguars (Panthera onca) rarely, if ever, attack people, but they do contribute to livestock depredation in South America (Rabinowitz 2005; Constant 2014). Attacks on people by mesocarnivores are rare, but these animals are infamous for predating on smaller stock and causing large-scale damages (Loveridge et al. 2010). In France, the Eurasian lynx (Lynx lynx) killed livestock from (Stahl, Vandel, Herrenschmidt and Migot 2001). The Eurasian lynx and caracal (Caracal caracal) are the only smaller felids (10 40 kg) that have been credited with regular livestock kills, with most responsible felids weighing > 50 kg (Inskip and Zimmerman 2009; Loveridge et al. 2010). Dholes (Cuon alpinus), dingoes (Canis lupus dingo), coyotes (Canis latrans) and even African wild dogs (Lycaon pictus) are responsible for livestock depredation in certain areas of the world (Wang and Macdonald 2006; Gusset et al. 2009). Some canid species have even been credited with accelerating the decline of certain threatened species, such as the Arctic fox (Vulpes lagopus) preying on nesting seabirds (Sillero-Zubiri and Switzer 2004) Livestock predation: The problem When examining HCC and proposing possible solutions it is vital to understand all aspects involved (Woodroffe et al. 2005). Carnivores have high dietary requirements and thus often display an opportunistic feeding behaviour (Inskip and Zimmerman 2009). Human populations are encroaching upon wildlife areas which have led to an increase in agricultural activities in these areas (Li et al. 2013; Constant 2014). People are living closer to protected areas and the buffer zones between human settlement and formally protected areas are becoming smaller (Gusset et al. 2009). In some instances people live in protected areas and even practice livestock farming inside the boundaries of the protected area (Pettigrew et al. 2012; Li et al. 2013). This could lead to the ungulate biomass in these protected areas consisting primarily of livestock (Bagchi and Mischra 2006). The term problem animal has been used to describe wildlife that has a negative effect on human lives (Linnell et al. 1999; Marker and Dickman 2005). Linnell et al. (1999) defined two different types of problem animals. The first being an animal which might just be in the wrong place at the wrong time, moving between ranges, dispersing from a protected area or where people wandered into 4

28 Chapter 1: Literature Review and Study Area their territories. The second type of problem animal is one that has a preference for livestock (or humans), usually more than other individuals of the same species. Male individuals have also been identified as the main culprits, primarily due to their larger home ranges, larger dispersal rates and possible larger body size (Linnell et al. 2001; Bunnefeld et al. 2006; Loveridge et al. 2010). In certain areas where conflict occurs the people in the region can also be at fault if, for example, no management practises are in place and animals are not protected from carnivores in any way (Inskip and Zimmerman 2009). Many people affected by human-carnivore conflict tend to turn to lethal persecution of carnivores, however, many of the species which are labelled as problem animals are also threatened species and protected legally from any lethal action (Treves, Wallace and White 2009; Thorn et al. 2013). Local people become more frustrated by such policies and in many cases still turn to illegal persecution of carnivores (Thorn et al. 2013; Constant 2014). Retaliatory killings by the affected human populations can result in a large-scale negative impact on the environment (Linnell, Swenson and Andersen 2001). There are areas where carnivores have been exterminated because it was thought that these animals were continuously feeding on livestock (Treves et al. 2004; Li et al. 2013). An example of this was the killing of wolves in the United States in response to the killing of livestock by these animals (Treves et al. 2004). In Bhutan, dholes were also nearly extirpated due to depredation on livestock (Wang and Macdonald 2006). In certain parts of Europe most of the larger carnivores have been exterminated or populations reduced in size due to conflict with people (Breitenmoser et al. 2010). The Eurasian lynx is one of the carnivores that were affected by human encroachment into natural areas in Europe (Stahl et al. 2001). In Scandinavia and Eastern Europe only a small number of individuals remained by 1900 and by 1940 in Western Europe, Eurasian lynx populations were declared extinct (Breitenmoser, Breitenmoser-Würsten and Capt 1998). Lynx have been reintroduced to Western Europe since 1970, however, conflict still remains and these animals are still being killed illegally by game hunters (Breitenmoser et al. 1998; Stahl et al. 2001). In the past, many countries, including South Africa, have had bounty systems in place to help control problem animals with the help of lethal methods (Beinart 2003; Musiani et al. 2005). In some cases the lethal control of a specific damage-causing animal can be justified (Anderson 1981; Loveridge et al. 2010). However, it can be challenging to identify the responsible problem animal and often the traps used are indiscriminate and many non-target individuals of a variety of species are also killed (Linnell et al. 1999). There are lethal methods that are more selective to the true problem animals. These include: 1) poison collars placed on livestock (usually a vulnerable individual), or 2) traps baited with recent kills by the problem individual (Burns, Zemilcka and Savarie. 1996). There are also many alternate, non-lethal approaches to manage carnivores preying 5

29 Chapter 1: Literature Review and Study Area on livestock (Treves et al. 2009). One of the more effective methods is the increase in traditional animal husbandry methods which includes guarding of livestock (Graham, Beckerman and Thirgood 2005). There is a difference in methods and approaches when implementing conservation strategies in developed versus non-developed countries (Loveridge et al. 2010). In some developed countries compensation schemes have been put in place to financially assist farmers that lose livestock to carnivores (Maclennan, Groom, Macdonald and Frank 2009; Treves et al. 2009). This requires an understanding of the conservation importance of carnivores and a strong commitment from both the private and governmental sectors (Constant 2014). However, draw-backs with compensation schemes also exist and Boitani, Ciucci and Raganella-Pelliccioni (2010) found that wolf-damage compensation programs in Italy were an unsustainable strategy to mitigate human-wolf conflict. In many developing countries governments do not have the resources to compensate for livestock losses and thus rarely support farmers (Dickman et al. 2011; Thorn et al. 2013). The lack of monetary support adds to the frustration experienced by local people regarding their livelihoods versus carnivores and leads to retaliatory killings of carnivores and often even non-target animals (Treves et al. 2006; Loveridge et al. 2010; Dickman et al. 2011). A trade-off also exists between promoting the economic growth of poorer countries by increasing agricultural activities with international opinions endorsing the conservation of threatened species (Treves et al. 2006). The attitudes of people towards the carnivores they come into conflict with can be damaging and increase the challenges of aiding mitigation of human-carnivore conflict (Dickman 2005; Schumann, Walls and Harley 2012). However, in certain instances the cultural or aesthetic importance of carnivores to local people might aid mitigation (Loveridge et al. 2010). Economic factors can also help predict the levels of conflict which can result from human-carnivore interactions (Bagchi and Mishra 2006). Historically wolves were intensely persecuted due to depredation on livestock; however conservation of wolves in North America has recently been intensified with scientific literature and non-scientific arguments claiming the importance of wolves in ecosystems (Mech 2011; Redpath, Gutiérrez, Wood and Young 2015). In many areas livestock farming provide the main income and many people depend on the sale of livestock for their and their families livelihoods. This is especially true for low-income regions where people practice subsistence farming (Thorn et al. 2013). When these farmers loose only one individual from their stock it has a higher negative impact on their livelihood than large-scale farmers that have more stock (Loveridge et al. 2010). In Australia, dingoes are responsible for the great decline in Australia s sheep flock (Letnic, Ritchie and Dickman 2012). It has also been suggested that livestock depredation by dingoes, along with feral dog hybrids, 6

30 Chapter 1: Literature Review and Study Area could be responsible for the collapse of the Australian sheep industry within the next years (Allen and West 2013). However, Forsyth et al. (2014) argue that these numbers are exaggerated. They do however agree that dingoes reduce the profitability of sheep farming, but mostly in rural communities where people do not have the resources to prevent dingo predation on their livestock. In Brazil, large ranches lost fewer cattle to carnivore attacks than smaller ranches (Michalski, Boulhosa, Faria and Peres 2006). Larger scale farming activities usually produce more money allowing the farmers to implement methods to decrease depredation (Loveridge et al. 2010; Thorn et al. 2013). The damage caused by some carnivores has been estimated, in monetary terms, to aid the understanding of how this negatively affects people living in some of these conflict areas (Ogada, Woodroffe, Oguge and Frank 2003; Patterson et al. 2004; Van Niekerk, Taljaard and De Waal 2013). In Kenya and Zimbabwe, where most farmers are subsistence farmers, 11% to 12% of annual income is lost to predation on livestock, primarily by lions and leopards (Ogada et al. 2003). A study conducted by Madhusudan (2003) found that households living in or in close proximity to Bhadra Tiger Reserve in India lost an estimated 12% of their livestock to tigers. In some cases stock owners lose more individuals to disease, theft or even starvation (Constant 2014). In the Kweneng district, Botswana, only 0.34% of livestock losses were attributed to carnivores. Losses to other factors such as disease, theft, accidents or malnutrition were much higher, accounting for % of total stock loss (Schiess-Meier et al. 2007). However, there are some areas such as Bhadra Tiger reserve where livestock losses were 4 times more than losses from other causes such as disease (Madhusudan 2003; Constant 2014). It is true that carnivores do have an adverse effect on livestock numbers in certain areas of the world; however it is always important to note that other factors can also contribute to the decline in stock numbers (Dar et al. 2009; Constant 2014; Forsyth et al. 2014). Though conservation organisations do try and resolve conflict where possible, some local people take offense to this action and feel robbed of their responsibility to look after their own stock (Treves, Wallace, Naughton-Treves and Morales 2006). When their safety is being jeopardised and economic losses are high there will be little co-operation with conservation strategies, as it is not seen as a priority (Goodrich 2010; Pettigrew et al. 2012). Conservation strategies are thus of great importance and human-carnivore conflict situations should be approached with caution and sensitivity (Loveridge et al. 2010; Suryawanshi, Bhatnagar, Redpath and Mishra 2013). Many times conservation strategies to mitigate human-wildlife conflict fail as a result of continued hostility towards wildlife from people or from other conservation conflicts which emerge during planning or implementing of plans (Treves et al. 2006; Dickman 2010). Conservation conflicts pose a challenge 7

31 Chapter 1: Literature Review and Study Area whenever a sustainable mitigation plan has to be implemented. Such conflicts include: 1) conflict of interest where stakeholders argue over the use of an area or resource, 2) conflicting beliefs and values, 3) conflicts over processes applied and the implementation of such solutions, 4) scientific studies are conducted, sometimes with little cooperation and input from locals, 5) conflicts regarding legislation and equality of different stakeholders and 6) personal conflicts between stakeholders (Redpath et al. 2015). In developing countries where the majority of carnivore persecution occurs, there have been various conservation efforts to help decrease livestock losses and mitigate conflict (Treves et al. 2006), which includes the provision of solutions to decrease predation. In some cases translocation of problem animals has been used, however, it has been found that translocation does not always work for felids (Athreya, Odden, Linnell and Karanth 2010). Leopards have been recorded returning to their previous range (Stander 1997; Athreya 2006), along with other felids such as jaguars (Rabinowitz 1986) and cougars (Ruth et al. 1998). The use of livestock guarding dogs over the years has proved to be a successful solution (Gehring, Vercauteren and Landry 2010; Marker, Dickman and Schumann 2005). When using livestock guarding dogs to control depredation by coyotes in North America, an 11% decrease in livestock predation was observed (Smith, Linnell, Odden and Swenson 2000). Livestock guarding dogs are effective, but commitment is needed for this method to work as effectively as possible (Smith et al. 2000; Marker et al. 2005). Additionally, in some cases, llamas, donkeys and even domesticated buffalo have been found to be effective livestock guards (Crawshaw 2004). Herders (or shepherds) have also been found to be an effective livestock guarding practice (Frank, Woodroffe and Ogada 2005; Loveridge et al. 2010). In developing countries these roles are often filled by younger family members, preventing them from attending school, thereby driving the poverty circle even more (Dickman et al. 2011). In some cases herders are employed, providing a work opportunity, however these individuals might be at risk of diseases, be sleep deprived or even fall prey to carnivores themselves (Barua, Bhagwat and Jadev 2013). In some developed countries many traditional animal husbandry methods have been abandoned due to a more recent lack of predators occurring in these regions (Stahl et al. 2001). However, with the reintroduction and in a few cases the reoccurrence of these carnivores in certain areas, many people have found themselves losing livestock having lost their knowledge of animal husbandry methods, such as shepherds and guarding dogs (Stahl et al. 2001). Kraaling (enclosures, bomas) has been abandoned in some areas as these structures lead to overgrazing and land degradation, making the area more susceptible to erosion (Beinart 2003). The use of kraals have also been criticised as one of the reasons why some felids practice surplus killings (Nowell and Jackson 1996). However, Schiess-Meier et al. (2007) state that only a small percentage of livestock killings in Botswana occurred in an enclosure where animals 8

32 Chapter 1: Literature Review and Study Area were guarded at night. In conclusion, the best method that has been developed over the years includes a combination of securely built enclosures, guard dogs and herders/shepherds (Ogada et al. 2003) Human-carnivore conflict in South Africa: with a focus on livestock losses Historically, livestock farming has been a part of human lives in South Africa potentially before the 17 th century (Beinart 2003). Before colonisation farming practices are thought to have been mostly nomadic, with high-intensity livestock guarding practices due to the large number of free-ranging predators. With the arrival of Jan van Riebeeck in 1652 and European settlers developing land for habitation, human-carnivore conflict intensified in the Cape (Stadler 2006). Dutch settlers started to farm on a large-scale with stock (sheep mostly obtained from the Khoisan people) and were dissatisfied with large predators causing damage to their stock (Beinart 2003). In the Cape Province, by the mid-nineteenth century, larger predators such as lions and hyenas (Crocuta crocuta) were extirpated from the area (Van Sittert 1998). Game species, such as the famous quagga (Equus quagga quagga) and bluebuck (Hippotragus leucophaeus), were also hunted to extinction for food and the decrease in natural prey for predators is cited as one of the main reasons why predators shifted to livestock predation (MacKenzie 1988; Brassine 2011). Predators such as leopards and caracals which also inhabit mountainous terrain used these areas to seek refuge in addition to restricting peak activity to night time (Beinart 2003; Skead 2011). Both these animals remained a problem for farmers, along with other canid species such as the black-backed jackal (Canis mesomelas), which increasingly started to inhabit areas where large predators were no longer present (Beinart 2003; Stadler 2006). The ability of black-backed jackals to adapt to changing circumstances and altering their ecology to compensate for disturbances have made these animals one of the most problematic animals in livestock farming in South Africa (Bekoff et al. 1984; Beinart 2003). Historical reports have shown that wild dogs were a considerable problem for Dutch settler farmers all the way back to the seventeenth and eighteenth century (Stadler 2006; Skead 2011). R. G Cumming (1856) claimed that wild dogs were known for not only killing stock which could be fed on, but killing any stock that crossed their path (Beinart 2003). Presently these animals are endangered with only a couple of small populations still remaining in South Africa (Gusset et al. 2009). Today, leopards and to some extent caracals are known for surplus killing where a large amount of animals are killed in one night, but only few are fed on (Bothma and Walker 1999; Marker and Dickman 2005). 9

33 Chapter 1: Literature Review and Study Area Predators such as black-backed jackals and caracal were later classified by the government as vermin or ongedierte (in Dutch/Afrikaans) [Beinart 2003; Stadler 2006]. Bounties were put in place to control jackal and other problem animal numbers, and so called poison clubs and dog hunting clubs were established (Van Sittert 1998). The South African government subsidised predator control strategies prior to 1990, due to the large number of losses experienced by farmers (Bergman et al. 2013). In 1887 when the poison clubs were started, this practice was subsidised by the Department of Agriculture (Du Plessis 2013). Farms were not fenced, so many farmers implemented the kraaling system to protect livestock from predators, however this system was cause for land degradation and many people blamed these farmers for the desertification of some areas of South Africa (Beinart 2003; Van Niekerk et al. 2013). Kraaling of animals was also cause for increased disease transmission between animals (Van Sittert 1998). By the 1910 s, carnivores were responsible for a stock losses of between 5-12% annually (Beinart 2003). This was the same percentage of stock that at that time was sent to abattoirs to be slaughtered for human consumption (Beinart 2003). In 1912 the government passed the Fencing Act (1912) and provided monetary support for farmers to fence their property, as well as providing mechanisms to facilitate the cooperation of neighbours when constructing fences (Van Sittert 1998; Beinart 2003; Bergman et al. 2013). In 1965, all smaller predator hunting concessions were abolished and merged into one hunting organisation, namely Oranjejag (Ferreira 1988). Oranjejag was run by government subsidies and all livestock farmers were expected to be members (Du Plessis 2013). Since the 1990s the South African government has ceased any subsidies to farmers for predator control and each farmer is now responsible for predator management on their specific farms (Du Plessis 2013). While government was supporting predator management strategies, a large number of predators were killed. From a total of caracals were reportedly killed and later, from , about caracals were killed annually only in the Karoo region (Marker and Dickman 2005; Bergman et al. 2013). Black-backed jackal were controlled even more intensely with over animals reportedly killed in the Cape Province from (Van Sittert 1998). With such large historical losses it is surprising that these predators still persist in large parts of South Africa and additionally are still considered problem animals (Brassine 2011; Nattrass and Conradie 2013). Problem animals or damage-causing animals are defined under the National Environmental Management Biodiversity Act, Act 10 of 2004 (NEMBA) (SA) as a wild vertebrate animal which causes damage to stock or other wild specimens, damages crops, natural flora or private property, threatens human life and occurs in such high numbers that agricultural grazing is being depleted [Department of Environmental Affairs 2010; Brassine 2011]. Very little research has been conducted in South Africa in order to fully understand predation on livestock by predators (Du Plessis, Avenant 10

34 Chapter 1: Literature Review and Study Area and De Waal 2015). Van Niekerk (2010) estimated that R1.39 billion is lost per annum due to predation of livestock. In South Africa many farmers have also shifted from livestock farming to game farming in an attempt at better financial security. However, many predators also predate on wild ungulates, especially fawns, so the conflict between predators and famers continues (Marker et al. 2005; Thorn, Green, Dalerum and Bateman 2012; Thorn et al. 2013). Currently predators are controlled on farms, both legally and illegally, using various lethal measures such as shooting, poisoning and the use of traps (Van Niekerk 2010; Forbes 2011; Du Plessis 2013). These strategies have thus far not been as successful as hoped. Conradie and Piesse (2013) found that leopard and caracal culling lead to increased livestock losses the following year. Many predators also travel vast distances to occupy empty territories, resulting in constant recruiting (Norton and Lawson 1985; Balme et al. 2009; Hayward and Kerley 2009). Black-backed jackals are also known to adapt their reproductive strategy, in reaction to increased persecution, by increasing litter sizes and by breeding at a younger age (Bingham and Purchase 2002; Beinart 2003; Nattrass and Conradie 2013). The killing of a dominant black-backed jackal results in sub-adults moving into the vacant territory which can result in smaller home ranges and a larger density of jackals (Bothma 2002; Ray, Hunter and Zigouris 2005; Brassine 2011). The lack of knowledge on the ecology of predators on farmlands in South Africa is surprising, as these problems have been part of farming for generations (Du Plessis 2013; Van Niekerk et al. 2013; Du Plessis et al. 2015) Importance of diet in carnivore studies and mitigating HCC Carnivores play an important role in any ecosystem and are considered important drivers of ecosystem structure and function (Beschta and Ripple 2009). Large carnivores exhibit a top-down regulatory role, not only on other smaller carnivores, but also on herbivore biomass (Prowse et al. 2014; Newsome et al. 2015). Medium-sized to smaller carnivores play important roles in ecosystem functioning and the loss of such carnivores can be detrimental to an ecosystem (Estes et al. 2011; Bagniewska and Kamler 2013). By preying on smaller prey, such as invertebrate species, birds and rodents, these carnivores also play a vital role in pest control on agricultural landscapes (Blaum, Tietjen and Rossminth 2009). Despite the importance of carnivores in natural systems their numbers are still decreasing worldwide, due to anthropogenic factors (Kissui 2008; Inskip and Zimmerman 2009). The availability of prey is one of the most important factors affecting the diet of a carnivore and this in turn influences the morphological, behavioural and physiological adaptations of that carnivore (Swanepoel et al. 2012; Kok and Nel 2004). Mammalian carnivores are vulnerable to extirpation due to their usually low densities and their large spatial requirements (Cardillo et al. 2005). Even though protected areas are critical to ensure the long-term persistence of carnivores in 11

35 Chapter 1: Literature Review and Study Area ecosystems (Balme, Slowtow and Hunter 2010) it is also important to have effective conservation management strategies for non-protected areas as a large amount of predator habitat constitutes areas outside of reserves (Martins 2010; Swanepoel et al. 2012). Most studies on carnivore diet have focussed in protected areas. However, such dietary studies do not confront the pressing issue of HCC in a changing world (Balme, Lindsey, Swanepoel and Hunter 2013). Many carnivores are known to change their ecology to adapt to changing surroundings (Woodroffe 2000; Van de Ven, Tambling and Kerley 2013). These adaptable species might not be as vulnerable as some specialist carnivores, but their adaptive behaviour has resulted in HCC in many areas (Kamler et al. 2012b; Van Niekerk et al 2013). Food availability is one of the drivers determining the persistence of free-roaming carnivores, a factor which has often led to these animals depredating on livestock (Balme, Hunter and Slowtow 2007; Loveridge et al. 2010). It is thus crucial to understand what the feeding ecology of carnivores is outside of reserves where they continue to live in close proximity to humans and to what extent carnivore feeding ecology may have changed (Balme et al. 2009; Suryawanshi et al. 2013; Thorn et al. 2013). One of the main reasons for livestock depredation is the decrease in wildlife prey species and an increase in available domestic prey species (Pettigrew et al. 2012). Focussing on both the relevant carnivores and prey availability is crucial to understanding the drivers behind the feeding behaviour of these animals (Thorn et al. 2013). Such strategies can provide a better understanding of the ecology of these animals and lead to important solutions to mitigate HCC. Worldwide many conflict mitigation studies have focused on dietary studies for baseline information to guide further research (Meriggi and Lovari 1996; Cunningham, Gustavson and Ballard 1999; Bacon, Becic, Epp and Boyce 2011). Morehouse and Boyce (2011) studied wolf diet in North America using a combination of GPS (Global Positioning System) cluster visitations and scat analysis and found cattle to be a prominent prey item. This was contradictory to previous studies which found primarily wild ungulates to occur in wolf diet. The aforementioned study also emphasised the importance that seasonality and land-use inclusion can have on a carnivore s diet, especially when the species is thought to play a role in livestock depredation (Morehouse and Boyce 2011; Du Plessis et al. 2015). Azevedo (2008) studied puma and jaguar diet in the Iguaçu National Park Area, South Brazil. Jaguars are said to be responsible for frequent livestock losses in this area due to a high abundance of livestock just outside of the park boundary and as a result of heavy persecution the Park now contains the last remaining population of jaguars in Southern Brazil (Conforti and Azevedo 2003). Azevedo (2008) assessed the diet of two carnivores in the area and found that jaguars predated on livestock to a higher extent that pumas. This study also concluded that livestock was 12

36 Chapter 1: Literature Review and Study Area only an alternative prey source for jaguars. In the instances where farms border protected areas, conflict between predator and farmer seems to be higher (Dar et al. 2009; Thorn et al. 2013). Many predators move into surrounding farmlands and often come across livestock, which is an abundant food source in these areas (Gurung and Seeland 2008; Balme et al. 2009). Similar to Azevedo (2008), Rowe-Rowe (1983) also found livestock remains in analysed black-backed jackal scats collected in a protected area in Kwazulu-Natal, South Africa. There is a high possibility that jackal use protected areas as refuge sites to escape persecution (Kaunda 2001; Kaunda and Skinner 2003; Loveridge and Macdonald 2004). The above mentioned studies were able to provide baseline strategies and management recommendations to prevent livestock losses, as well as providing information to local farmers regarding the extent to which these animals rely and predate on livestock (Inskip and Zimmerman 2009; Thorn et al. 2013). Historically South Africa had widespread populations of free-ranging large carnivores (Skead 2011). Leopards are the last remaining apex predators in many small-stock regions, similar to Namaqualand, Northern Cape (Martins 2010; Skead 2011). However, studies on the role of these apex predators, which include livestock farms in the study area, are limited. In 1981, Stuart analysed the diet of various carnivores in the Cape Province. Leopard diet was analysed from 36 stomachs collected at 30 different localities across the Cape. However, stomachs were collected during control operations and the high occurrence of livestock recorded in the diet was seen as an overestimation due to sampling bias (Stuart 1981). Norton, Lawson, Henry and Avery (1986) were one of the first studies on leopard diet in the Western Cape and included a large scope of land-uses. Martins et al. (2011) studied leopard diet in the Cederberg Conservancy, Mann (2014) studied leopard diet in the Little Karoo, also in the Western Cape and Braczkowski, Watson, Coulson and Randall (2012a) in the Southern Cape. All three of the latter studies assessed leopard diet in a matrix of different land-uses and found stock to only contribute a small percentage to the total diet of leopards. In South Africa livestock losses from carnivores such as black-backed jackal and caracal are thought to be high (Bergman et al. 2013; Du Plessis et al. 2015). However, most dietary studies on these two mesocarnivores have been done in the confines of protected areas, providing very little information as to the real impact of these animals on livestock farms (Du Plessis et al. 2015). The caracal, a smaller felid, is thought to be responsible for more livestock losses than leopard on small stock farms (Thorn et al. 2013). Past studies on caracal and black-backed jackal diet, mostly focussing on stomach contents, included small stock farms in their study area. However, recent studies are limited and studies in the Succulent Karoo biome are lacking (Du Plessis et al. 2015). Moolman (1984) studied caracal diet in the Mountain Zebra National Park (MZNP) and surrounding farmlands 13

37 Chapter 1: Literature Review and Study Area in the Cradock region, Eastern Cape. Caracal scats collected on the farmlands contained a high percentage of stock remains. Stuart (1981) found domestic stock to occur most frequently in caracal diet from 194 stomachs analysed from 135 localities in the Cape Province. Domestic stock was the second most frequently consumed prey item in black-backed jackal diet as analysed from 143 stomachs from 65 localities. Once again, the majority of the stomachs analysed were obtained from control operations, thus providing inadequate results (Stuart 1981). A more recent study on caracal diet which includes a variety of land-uses was done by Braczkowski et al. (2012b) which found a very low occurrence of livestock in caracal diet in the Southern Cape study area. However, farms included were mostly cattle farms providing very little novel insight to caracal predation on small-stock farms. A recent black-backed jackal study by Kamler et al. (2012a) found a seasonal variation in jackal diet, coinciding with the lambing periods. However, this study was carried out on only one small stock farm and jackal diet is known to vary spatially and temporally (Pyke, Pulliam and Charnov 1977; Brassine 2011). Du Plessis et al. (2015) assessed past research on both caracal and black-backed jackal ecology and the relevance of studies to human predator conflict management. The authors expressed concern for the lack of available relevant scientific knowledge. Research required on black-backed jackals and caracal included, 1) increased knowledge of these two mesocarnivores territoriality, densities and ranging behaviour on livestock farms, 2) prey selection and timing of predation since it is clear that black-backed jackals and caracals exhibit an opportunistic feeding behaviour, but unclear whether some individuals may have developed a specialisation towards livestock predation, 3) timing of reproduction and whether it can coincide with lambing periods and 4) the controversial issue of compensatory breeding where it is believed that reproduction rates, litter sizes and age of sexually matured individuals might be adapted to compensate for an increased persecution of the species (Du Plessis et al. 2015) Focal Species The leopard (Panthera pardus) an apex predator The leopard (Panthera pardus) is the most widely distributed large felid in the world particularly in Africa, where the highest numbers of leopards are currently found (Nowell and Jackson 1996; Skinner and Chimimba 2005). The success of leopards throughout such a wide range (occurring throughout sub-saharan Africa, in the Middle East and parts of Asia) can mostly be attributed to their secretive nature and opportunistic feeding behaviour (Balme et al. 2007; Estes 2012; Chattha et al. 2015). Leopards occupy a wide variety of habitats ranging from semi-deserts to forested areas. In the tropical forests of Africa leopards are the only large predator still persisting (Estes 2012). 14

38 Chapter 1: Literature Review and Study Area Leopards have been found to be more successful than other larger carnivores, such as hyena and lions, when living in close proximity to humans (Kissui 2008). The recorded preferred habitat for leopards is rocky outcrops (or koppies), hills, mountain ranges and forests; habitat types which allow for cover and refuge (Skinner and Chimimba 2005). Leopards have distinct markings in the form of rosettes and no two individuals will be found with the same physical features (Estes 2012). Weighing between 20 and 90 kg, leopards have a variable body mass further supporting an opportunistic feeding behaviour across a wide range of habitats (Hayward et al. 2006). A clear difference in size exists between males and females, with males generally being larger than females (Skinner and Chimimba 2005; Balme, Hunter and Braczkowski 2012a). Leopards are solitary felids, only associating with another individual long enough to mate (Estes 2012). Balme et al. (2012b) found that 40% of cub deaths were attributed to infanticide. Leopards are also territorial predators with males generally holding larger territories than females (Skinner and Chimimba 2005; Martins 2010). Leopards are nocturnal, thus mostly hunting and moving around at night (Estes 2012). Some areas where cover is in excess, such as forested habitats, leopards may exhibit crepuscular and even diurnal behaviour (Martins and Harris 2013). Utilising stalking behaviour and being ambush hunters, leopards mostly rely on cover to conceal their movements (Balme et al. 2007; Estes 2012). The distances stalked by leopards vary depending on habitats (Stander et al. 1997). This reliance on cover as part of their hunting strategy can limit leopards to mostly remain in areas with adequate cover (Hayward et al. 2006). As mentioned these animals are extremely opportunistic in their feeding behaviour and in sub-saharan Africa alone, 92 prey species have been recorded for leopards (Balme et al. 2007). Due to their body mass leopards require 1.6 to 4.9 kg of meat per day (Bothma and le Riche 1986; Stander et al. 1997; Hayward et al. 2006). Leopard diet mostly includes prey items weighing between kg (Hayward et al. 2006) and can prey on anything ranging from invertebrates to an adult eland (Taurotragus oryx) [Bailey 1993; Hayward et al. 2006; Martins et al. 2011]. In 2008 the conservation status of leopards was reassessed by the International Union for Conservation of Nature (IUCN) and a decision was made to change their conservation status from Least Concern to Near Threatened (Henschel et al. 2008; Chattha et al. 2015). This reassessment was made due to the fact that despite leopards being common in certain areas their numbers are still decreasing over the extent of their range (Ray et al. 2005). A dramatic reduction of leopard numbers has been observed in Africa, where leopard range has been reduced by 37% (Ray et al. 2005; Balme et al. 2010). Some leopard species such as the Sri Lankan leopard (Panthera pardus kotiya), and the Persian leopard (Panthera pardus saxicolor), are already classified as Endangered 15

39 Chapter 1: Literature Review and Study Area with the Amur leopard (Panthera pardus orientalis) and the Arabian leopard (Panthera pardus nimr), classified as Critical (Henschel et al. 2008). In 1986 it was estimated that only a mere 13% of potential leopard range was within the boundaries of protected areas (MacKinnon & MacKinnon 1986; Balme et al. 2010). See Figure 1.1 for a map illustrating remaining suitable leopard habitat in South Africa. Figure 1.1. Suitable leopard habitat still available in South Africa according to Swanepoel et al. (2012) developed from a model with various environmental variables, a) excluding human variables, b) estimated human impact and c) where habitat suitability index represents logistic probabilities of occurrences. Negative values indicate areas where human impact has a negative effect on leopard habitat, positive values where human impact had a positive effect and zero values where leopard habitat suitability was not influence by human impacts The mesocarnivores The caracal (Caracal caracal) The caracal (Caracal caracal) is one of the most widespread felids on the African continent (Avenant and Nel 2002; Skinner and Chimimba 2005), occurring in the entire Southern African region extending to the margins of the Sahara Desert in the North, Morocco, Egypt, Sudan, Ethiopia, Somalia, Mauritania and northern Niger (Stuart 1982, Smith 2012). Caracal are also found in the Middle East, Eastern Turkey, the Arabian Peninsula, Turkmenistan, Pakistan, India, Kazakstan, Afganistan, Tajikistan and Uzbekistan (Stuart 1982; Nowell and Jackson 1996). They occupy a wide range of habitats including arid areas, open savannas, open grasslands and also the Afromontane and evergreen forests in South Africa and the tropical forests of the Democratic Republic of Congo (Stuart and Wilson 1988; Skinner and Chimimba 2005; Estes 2012; Smith 2012). Sometimes referred to incorrectly as a lynx, these animals were first classified under the genus Felis (Skinner and 16

40 Chapter 1: Literature Review and Study Area Chimimba 2005; Smith 2012). However, it was re-classified in the genus Caracal with one other felid, Caracal aurata, the African golden cat (Wozencraft 1993; Morales et al. 2003). Caracal received their name from the Turkish word garah-gulak or caracal when translated to english, which means black-eared (Skinner and Chimimba 2005; Ghoddousi, Ghadirian and Fahimi 2009). Caracal used to be captured and trained to hunt for people in India and Iran (Divyabhanusinh 1995; Ghoddousi et al. 2009). Caracal are solitary cats, mostly active at night, but diurnal activity has been recorded (Avenant and Nel 1998; Iliman and Gürkan 2010; Estes 2012). These medium-sized cats, the largest of the smaller felids, can weigh up to 12 kg (females) and 15 kg (males) [Skinner and Chimimba 2005]. Males also hold larger territories than females, with more than one individual s home range overlapping with another (Stuart 1982). Studies in the Western Cape, South Africa, also discovered that caracals can travel vast distances before settling in a specific area (Norton and Lawson 1985; Bothma and Le Riche 1994). Norton and Lawson (1985) tracked a young male caracal which moved around in an area of 483 km² before settling in an area of 65 km² for 11 months. One of the fastest felids, caracal catch their prey by means of a fast-paced dash and are known to propel themselves into the air to catch airborne prey (Estes 2012; Smith 2012). These felids are predominantly hunters, but have been observed to scavenge when resources are limited (Stuart and Hickman 1991; Skinner and Chimimba 2005). Caracals have persisted in areas with high fragmentation and human development. This could mostly be ascribed to their secretive nature and high use of areas with cover (Stuart 1982; Skinner and Chimimba 2005). Caracals also tend to feed on prey items that are available in high numbers. These felids are opportunistic hunters with a generalist diet, comprising mostly of mammals, but may also include birds, reptiles and arthropods (Palmer and Fairall 1988; Estes 2012). Previous studies have recorded a prey range extending from 1 g to 31 kg in mass (Grobler 1981; Moolman 1984; Palmer And Fairall 1988; Avenant and Nel 2002; Braczkowski et al. 2012b). Although widespread and relatively common, very little information is published on caracal feeding ecology (Stuart 1981; Avenant and Nel 1998, Braczkowski et al. 2012b). Caracals are reported to be one of the main carnivores responsible for small stock losses in South Africa, along with the blackbacked jackals (Canis mesomelas) (Bergman et al. 2013; Du Plessis 2013). Between 1931 and 1952 an average of 2219 caracals were killed every year in the Karoo region of South Africa to help control predator numbers (Stuart 1981). The caracal has been categorised by the International Union for Conservation of Nature (IUCN) as Least Concern (Breitenmoser-Wursten, Henschel and Sogbohossou 2008). Caracal population numbers seem to be stable across their distribution in Africa 17

41 Chapter 1: Literature Review and Study Area due to being widespread and common, however in parts of Asia there are concerns that populations are declining due to habitat destruction (Ray et al. 2005) The black-backed jackal (Canis mesomelas) The black-backed jackal (Canis mesomelas) can clearly be distinguished by the dark saddle on the upper parts of the body. This feature aids in distinguishing black-backed jackal from the side-striped jackal (Canis adustus), along with having a lighter mass than the side-striped jackal (Loveridge and Nel 2004; Skinner and Chimimba 2005; Estes 2012). Black-backed jackal are common in arid areas. They are distributed throughout most of Southern Africa, including Namibia, south-west Angola, Botswana, south-west to east Zimbabwe, and the most southern parts of Mozambique. The species is found throughout South Africa, except for highly urbanised areas and the forested regions of Knysna. Black-backed jackals also occur in the more northern parts of Africa, from the Gulf of Aden southwards into southern Tanzania, 900 km from the edge of its southern distribution (Skinner and Chimimba 2005). Black-backed jackal are specially adapted for survival in drier regions having kidneys with a thick medulla which allows black-backed jackals to concentrate their urine in times of drought (Loveridge and Nel 2004; Brassine 2011). These canids show a preference for open habitats, but have been recorded in a wide range of habitat types including Nama-Karoo, Succulent Karoo, open and arid savannah, fynbos, arid coastal deserts and grasslands (Loveridge and Nel 2004; Skinner and Chimimba 2005; Estes 2012). Black-backed jackals also occur on farmlands, being drawn to an abundance of potential prey. Black-backed jackals display both diurnal and nocturnal activity patterns (Estes 2012). In many protected areas where jackals are not persecuted or in areas of low human habitation they are often seen during the day, however in areas of high human activity and high persecution they are mostly active a night (Loveridge and Nel 2004; Skinner and Chimimba 2005). Black-backed jackals are infamous for their adaptable behaviour as a response to human activity (Skinner and Chimimba 2005). Most of the species' main prey items, such as certain rodent species are diurnal, another reason for their high activity during day (Ferguson, Galpin and De Wet 1988). Black-backed jackal are mostly seen travelling at a trot and generally only walk slowly when hunting for rodents and invertebrates with their ears pricked, listening for any prey activity (Skinner and Chimimba 2005). Like most canid species black-backed jackals are not solitary and have been observed to either forage singly, in pairs or at times in groups of three or more (Rowe-Rowe 1983). Little sexual dimorphism exists, however males are larger than females (Skinner and Chimimba 2005). In the drier western regions males exhibit a deep reddish brown coat in the winter months (Loveridge and Nel 18

42 Chapter 1: Literature Review and Study Area 2004). They are monogamous and a dominant pair will mark and defend their territory against intruders (Moehlman 1986; Ferguson, Nel and De Wet 1983; Skinner and Chimimba 2005). A dominate pairs territory will usually exclude other dominant pairs, however in other cases blackbacked jackal territories may overlap (Walton and Joly 2003; Skinner and Chimimba 2005; Estes 2012). Black-backed jackal hunting and scavenging activities have been observed in various ecosystems, especially in the savanna and open grassland areas (Owens and Owens 1978; McKenzie 1990). Although small in size, these canids are very proficient hunters (Lamprecht 1978; Estes 2012). Various studies have observed black-backed jackal forming groups to more effectively hunt larger antelope species (Estes 2012). Kamler, Foght and Collins (2009) observed a single adult black-backed jackal successfully killing an adult impala. Black-backed jackals are clear omnivores and exhibit a generalist diet (Loveridge and Nel 2004; Skinner and Chimimba 2005). Studies have found that mammals, insects, carrion and vegetable matter, such as seeds and fruits, constitute the largest portion of jackal diet (Rowe-Rowe 1976; Lamprecht 1978; Kok 1996; Nel, Loutit and Bothma 1997; Loveridge and Macdonald 2003; Do Linh San et al. 2009; Kamler, Klare and Macdonald 2012ᵃ). Past studies have reported high ungulate occurrence in black-backed jackal diet (Lamprecht 1978; Kok 1996; Do Linh San et al. 2009; Klare, Kamler, Stenkewitz and Macdonald 2010), however, most studies in diverse areas found rodents to be the dominant prey item (Rowe-Rowe 1983; Stuart 1987; Van der Merwe et al. 2009). Many authors have emphasised the complications with separating carrion and hunted prey remains from scat and stomach content analyses (Smithers 1983; Kok 1996). Smithers (1983) recorded > 50% of black-backed jackal diet being made up of insect remains from 96 stomachs analysed. Black-backed jackal are opportunistic hunters and scavengers and will mostly choose prey according to its high abundance, as well as selecting for prey that are easily captured (Skinner and Chimimba 2005). Black-backed jackals have a long history of conflict with farmers in South Africa (Beinart 2003). Human presence is one of the main reasons for a decrease in natural prey across Southern Africa. Being opportunistic foragers, jackals have benefitted from the increased prey biomass available in the form of livestock (Brassine 2011; Kamler et al. 2012a). The persecution by humans could have altered many aspects of jackal biology (Skinner and Chimimba 2005). Black-backed jackal populations are capable of recovering from stresses placed on them, such as persecution, by exhibiting compensatory breeding (Beinart 2003; Nattrass and Conradie 2013). Ferguson et al. (1983) reported polygamy being exhibited by an alpha male in response to increased lethal persecution. Often farmers use lethal persecution to try and control black-backed jackal numbers, 19

43 Chapter 1: Literature Review and Study Area not realising that removing a dominant jackal allows sub-adults to move into the newly vacated territory (Bothma 2002; Ray et al. 2005). Sub-adults may be less efficient hunters and may select easier prey to catch such as livestock (Linnell et al. 1999). Black-backed jackal are not only being persecuted as a result of depredation on livestock, but more recently also due to them having a negative effect on the wildlife and game industry. In 2010, SANParks (South African National Parks) culled 132 black-backed jackal in the Karoo National Park and 212 in the Addo Elephant National Park due to the suspicion that jackals were responsible for springbok and other antelope decline (Nattrass and Conradie 2013). Despite the heavy persecution by farmers, both in the past and currently, predation on livestock by black-backed jackal has not decreased and black-backed jackal are still abundant in South Africa (Loveridge and Nel 2004; Blaum et al. 2009). The black-backed jackal is listed in the category least concern by the International Union for Conservation of Nature (IUCN) and has the lowest threat and vulnerability score of Africa s predators (Ray et al. 2005; Hoffmann 2014). 1.5 Study Area Location and History Namaqualand is situated in the western region of South Africa and also extends into Namibia from the Orange River in the south-west to Lüderitz (Cowling, Esler and Rundel 1999). In South Africa Namaqualand covers approximately km² and extends from the Olifants River and Bokkeveld Mountains in the Western Cape, northwards towards Loeriesfontein in the Northern Cape, to just east of Vioolsdrif on the Orange River and to the west at Alexander Bay (Cowling et al. 1999; Desmet 2007) [See Figure 1.2]. Namaqualand can be divided into seven bioregions based on climate, physical environment and flora (Hilton-Taylor 1996; Desmet 2007). The study area lies on the western border of the Kamiesberg bioregion and the eastern border of the Hardeveld in the Northern Cape, South Africa. The study area includes the eastern section of Namaqua National Park (S E ) and surrounding farmlands to the north, east and south of the national park. The Namaqualand National Park is situated approximately 495 km from Cape Town. It was proclaimed as a national park in 1988 and was established as an extension of the original 930 ha Skilpad Wildflower Reserve (Van Rooyen 2002; van Deventer and Nel 2006). Only the eastern, mountainous area of the national park was included in the study area in order to decrease variation in the physical environment. The study area provided an ideal location to investigate the diet of three predators in a protected area and its surrounding small stock farmlands. The entire Namaqualand region includes 420 private farms, covering about 52% of the region (Benjaminsen et al. 2006). 20

44 Chapter 1: Literature Review and Study Area The San people were the first humans to utilise parts of Namaqualand, however they never stayed in the area permanently. The first people that settled in Namaqualand were the Khoi-Khoi, sometimes referred to as the Nama people, almost 200 years ago (Kostka 2005; Benjaminsen et al. 2006). In 1806 the British commando arrived in Namaqualand, enslaving many of the people who lived there (Kostka 2005). Many Namas were also used as labourers for the Afrikaans trekboere (travelling farmers) [Benjaminsen et al. 2006]. In 1878 the British rule allowed farmers, the former trekboere, to buy the land they were using under tenure from the Dutch East India Company (Kostka 2005). Most towns in Namaqualand started as mission stations and became refuge for the Nama-khoi people in the area (Boonzaaier 1996). Presently many descendants of the Nama-khoi live and farm in communal areas, which make up 30% of Namaqualand (Benjaminsen et al. 2006). Figure 1.2. A map of South Africa (insert) showing the location of the study area (marked as the grid) Climate Namaqualand is classified as a semi-arid, winter rainfall region (Cowling et al. 1999). For the greater part of Namaqualand rainfall is reliable, especially when compared to other arid regions (Desmet 21

45 Chapter 1: Literature Review and Study Area 2007). However, rainfall can vary throughout the region and ranges from 50 mm annually in the north-west to up to 400 mm in the Kamiesberg region (Cowling et al. 1999). More specifically the study area of this study, primarily made up of Namaqualand Klipkoppe Shrubland, receives a mean annual precipitation (MAP) of 160 mm, with some years receiving < 100mm annually. These drought periods either last one or two years (Mucina and Rutherford 2006). Rainfall is the highest in June (Figure 1.3), but mostly occurs from May to September (Desmet 2007). The average rainfall recorded at Skilpad (S E ) at an altitude of 683 m above sea level over 15 years was 340 mm (Namaqua National Park 2012). Figure 1.3. Average monthly rainfall for the study area over a 7 year period from (data from Skilpad in Namaqua National Park). Summers are hot and can reach mean maximum temperatures of 30 C, while temperatures can drop to 5 C in the winter months, specifically in June and July (Mucina and Rutherford 2006) [Figure 1.4]. The highest recorded temperature for 2014 was 38.8 C (26 February 2014) and the lowest was 2 C (7 July 2014). Frost can occur for 8 days a year, but varies from year to year (Mucina and Rutherford 2006). Some years snow has fallen on the highest peaks of the Kamiesberg, but this area was not 22

46 Chapter 1: Literature Review and Study Area included in this study (Namaqua National Park 2012). Mist is common in the autumn and winter months and is said to be one of the factors important in seed germination in this area (Cowling and Pierce 1999). Figure 1.4. Average monthly maximum and minimum temperature for the study area over a 7 year period from (data from Skilpad in Namaqua National Park) Geology and Soils The landscape of Namaqualand is characterised by granite gneiss (Kamieskroon gneiss). This creates a scene of dome-shaped hills with flatter valleys in between (van Deventer and Nel 2006; Desmet 2007). Rock size varies from medium to large (Figure 1.5) to prominent rock domes (Figure 1.6) [Mucina and Rutherford 2006]. Elevation ranges from 180 m in the far west (not included in the study area) to 300 m at Melkboom (in Namaqua National Park) and finally to 750 m at Skilpad. The eastern section of the study area, including Skilpad, lies at the foothills of the Kamiesberg. In the broader area of Namaqualand, three different soil types have been identified (Watkeys 1998). Soils in the study area ranged from sand to loam (Mucina and Rutherford 2006; Desmet 2007) and 23

47 Chapter 1: Literature Review and Study Area varied from lime-rich and shallow, from red to yellow in colour in the eastern sections to red, granite-derived colluvial soils more towards the west (Van Deventer and Nel 2006; Desmet 2007). In the western edge of the study area heuweltjies (circular termitaria) were observed (Desmet 2007). These heuweltjies create large, visible patches (± 10 m) in the soil consisting of a higher nutrient content than the surrounding area (Moore and Picker 1991). Figure 1.5. The landscape of Namaqualand, consisting of medium to large granite gneiss and flatter valleys. Corlé Jansen 24

48 Chapter 1: Literature Review and Study Area Figure 1.6. The large rock domes of granite gneiss which can be seen in the study area. Corlé Jansen Vegetation The study area forms part of the Succulent Karoo biome, one of only two semi-arid biodiversity hotspots in the world. Namaqualand makes up approximately a quarter of the Succulent Karoo and boasts 3500 flora species in 135 families and 724 genera, of which 25% is endemic to Namaqualand (Driver, Desmet, Rouget and Cowling 2003; Desmet 2007). The broader vegetation type, which is included in the majority of the study area, is Namaqualand Klipkoppe shrubland, which according to Mucina and Rutherford (2006) falls under Namaqualand Hardeveld. The area consists of open shrubland of up to 1m in height, comprising of dwarf to medium-sized shrubs. Aloe dichotoma var. dichotoma, or better known as the kokerboom (quiver tree), can be found on the north-facing slopes (Mucina and Rutherford 2006). Along the dry riverbeds Acacia Karoo is found. Another tree species commonly encountered in the study area was the rock-splitting fig (Ficus ilicina), which is found on rocks or boulders (Trail 2015). Important succulent shrubs found in the study area included Euphorbia decussate (melktou) and Euphorbia mauritanica (melkbos). Both these shrubs produce a toxic milky substance once a stem is broken, to protect it from herbivores (Esler, Milton and Dean 2006). Didelta spinosa and Leipoldtia schultzei both display flowers after good rains. Other important succulent shrubs occurring in the study area included Cotyledon cuneata, C. orbiculata var. orbiculata, Crassula atropurpurea var. 25

49 Chapter 1: Literature Review and Study Area watermeyeri, Othonna cylindrical, Pelargonium crithmifolium, Ruschia goodiae, Sarcocaulon crassicaule, Tetragonia fruticose and Zygophyllum foetidum (Mucina and Rutherford 2006). Tall shrubs such as Dodonaea viscosa var. angustifolia (sand olive) which grows in sandy soils, Putterlickia pyracantha and the commonly encountered Rhus undulata also occur in the study area (Mucina and Rutherford 2006). Small shrubs occur in the area with Galenia africana (kraalbos), a shrub found in overgrazed areas or abandoned ploughed fields (Esler et al. 2006). Other small shrubs included the distinct Eriocephalus microphyllus var. pubescens or commonly known as kapokbos, Berkheya fruticose, Hermannia disermifolia, Lebeckia sericea, the spiny Acanthopsis spathularis, Asparagus capensis var. capensis,, Eriocephalus brevifolius, Galenia fruticose, Selago divaricate and S. glutinosa (Mucina and Rutherford 2006). In winter and spring Namaqualand is transformed by mass floral displays, a popular tourist attraction (van Rooyen 2002; Botha, Cariick and Allsopp 2008). The occurrence of annual wildflower displays are often a result of human interferences such as old fields and potential overgrazing sites (van Rooyen 2002) [see Figure 1.7 and Figure 1.8 for a comparison of the same fields and area at Skilpad in two different seasons] It is in spring season when herbaceous plants and geophytes such as Tripteris amplectens, T. hyoseroides, Arctotis revoluta, Gazania leiopoda, Ursinia cakilefolia, Felicia bergeriana, Heliophila variabilis, Leysera gnaphalodes, Conicosia elongata and Oxalis obtuse and Senecio arenarius appear (Mucina and Rutherford 2006; van Rooyen, Henstock, van Rooyen and van der Merwe 2010). Figure 1.7. The fields in front of Skilpad, Namaqua National Park in the dry months (December- May). Corlé Janse 26

50 Chapter 1: Literature Review and Study Area Figure 1.8. The fields in front of Skilpad, Namaqua National Park in the wet months (June- November). Corlé Janse 1.6 Objectives of this study The main objectives of this study were to: 1. Determine the general diet of leopard, caracal and black-backed jackal across two land-uses, in addition to comparing diet between the two land-uses namely the Namaqua National Park and surrounding small stock farms. 2. Determine the relative abundance index (RAI) of available prey from data obtained from both camera traps and small mammal trapping and compare the RAI between the Namaqua National Park and surrounding farmlands. 3. Determine prey preference of the three species using diet data and abundance data from camera traps. 4. Test if caracal diet outcomes differ when using two different methodologies, namely scat analysis and GPS cluster visitations or kill sites. 1.7 References Allen, B. L., and West, P Influence of dingoes on sheep distribution in Australia. Australian Veterinary Journal 91: Anderson, J. L The re-establishment and management of a lion Panthera leo population in Zululand, South Africa. Biological Conservation 19:

51 Chapter 1: Literature Review and Study Area Athreya, V Is relocation a viable management option for unwanted animals? The case of the leopard in India. Conservation and Society 4: Athreya, V., Odden, M., Linnell, J. D., And Karanth, K. U Translocation as a Tool for Mitigating Conflict with Leopards in Human-Dominated Landscapes of India. Conservation Biology 25: Avenant, N. L., and Nel, J. A. J Home-range use, activity, and density of caracal in relation to prey density. African Journal of Ecology 36: Avenant, N. L., and Nel, J. A. J Among habitat variation in prey availability and use by caracal Felis caracal. Mammal Biology 67: Bacon, M. M., Becic, G. M., Epp, M. T., and Boyce, M. S Do GPS Clusters Really Work? Carnivore Diet from Scat Analysis and GPS Telemetry Methods. Wildlife Society Bulletin 35: Bagchi, S., and Mishra, C Living with large carnivores: predation on livestock by the snow leopard (Uncia uncia). Journal of Zoology 268: Bagniewska, J. M., and Kamler, J. F Do black-backed jackals affect numbers of smaller carnivores and prey? African Journal of Ecology 52: Bailey, T. N The African leopard: ecology and behavior of a solitary felid. New York: Columbia University Press. Balme, G. A., Batchelor, A., Woronin Britz, N., Seymour, G., Grover, M., Hes, L., Macdonald, D. W., and Hunter, L. T. 2012b. Reproductive success of female leopards Panthera pardus: the importance of top down processes. Mammal Review 43: Balme, G. A., Hunter, L. T. B., and Braczkowski, A. 2012a. Applicability of Age-Based Hunting Regulations for African Leopards. PLoS ONE 7: e Balme, G. A., Slotow, R., & Hunter, L. T Impact of conservation interventions on the dynamics and persistence of a persecuted leopard (Panthera pardus) population. Biological Conservation 142: Balme, G. A., Slotow, R., and Hunter, L. T. B Edge effects and the impact of non-protected areas in carnivore conservation: leopards in the Phinda-Mkhuze Complex, South Africa. Animal Conservation 13: Balme, G., Hunter, L., and Slotow, R Feeding habitat selection by hunting leopards Panthera pardus in a woodland savanna: prey catchability versus abundance. Animal Behaviour 74:

52 Chapter 1: Literature Review and Study Area Balme, G., Lindsey, P. A., Swanepoel, L. H., and Hunter, L. T. B Failure of research to address the rangewide conservation needs of large carnivores: leopards in South Africa as a case study. Conservation Letters 00: 1 9. Barua, M., Bhagwat, S. A., and Jadhav, S The hidden dimensions of human wildlife conflict: Health impacts, opportunity and transaction costs. Biological Conservation 157: Beinart, W The rise of conservation in South Africa: settlers, livestock and the environment. Oxford: Oxford University Press. Bekoff, M., Daniels, T. J., and Gittleman, J. L Life history patterns and the comparative social ecology of carnivores. Annual review of ecology and systematics 15: Bengis, R. G., Kock, R. A., and Fischer, J Infectious animal diseases: the wildlife/livestock interface. Revue Scientifique et Technique-Office international des e pizooties 21: Benjaminsen, T. A., Rohde, R., Sjaastad, E., Wisborg, P., and Lebert, T Land reform, range ecology, and carrying capacities in Namaqualand, South Africa. Annals of the Association of American Geographers 96: Bergman, D. L., De Waal, H. O., Avenant, N. L., Bodenchuk, M. J., Marlow, M. C., and Nolte, D. L The Need to Address Black-backed Jackal and Caracal Predation in South Africa. In: J. B. Armstrong and G. R. Gallagher, eds Proceedings of the 15th Wildlife Damage Management Conference South Carolina. Beschta, R. L., and Ripple, W. J Large predators and trophic cascades in terrestrial ecosystems of the western United States. Biological Conservation 142: Bingham, J., and Purchase, G. K Reproduction in the jackals Canis adustus Sundevall, 1846, and Canis mesomelas Schreber, 1778 (Carnivora: Canidae), in Zimbabwe. African Zoology 37: Blaum, N., Tietjen, B., and Rossminth, E. Impact of Livestock Husbandly on Small- and Medium-Sized Carnivores in Kalahari Savannah Rangelands. The Journal of Wildlife Management 73: Boitani, L., Ciucci, P., and Raganella-Pelliccioni, E Ex-post compensation payments for wolf predation on livestock in Italy: a tool for conservation? Wildlife Research 37: Bonnet, X., Naulleau, G., and Shine, R The dangers of leaving home: dispersal and mortality in snakes. Biological conservation 89: Boonzaier, E The Cape herders: a history of the Khoikhoi of southern Africa. South Africa: New Africa Books. Botha, M. S., Carrick, P. J., and Allsopp, N Capturing lessons from land-users to aid the development of ecological restoration guidelines for lowland Namaqualand. Biological Conservation 141:

53 Chapter 1: Literature Review and Study Area Bothma, J. D. P Game ranch management. South Africa: Van Schaik Publishers. Bothma, J. du P., and Le Riche, E. A. N Prey preference and hunting efficiency of the Kalahari Desert leopard. In: S. D. Miller and D. D. Everett, eds. Cats of the world: Biology, conservation and management Washington, D. C: National Wildlife Federation. Bothma, J. du P., and Le Riche, E. A. N Scat analysis and aspects of defaecation in Northern Cape leopards. South African Journal of Wildlife Research 24: Bothma, J. Du P., and Walker, C Larger Carnivores of the African Savanna. New York: Springer-Verlag. Braczkowski, A., Watson, L., Coulson, D., Lucas, J., Peiser, B., and Rossi, M. 2012b. The diet of caracal, Caracal caracal, in two areas of the southern Cape, South Africa as determined by scat analysis. South African Journal of Wildlife Research 42: Braczkowski, A., Watson, L., Coulson, D., and Randall, R. 2012a. Diet of leopards in the southern Cape, South Africa. African Journal of Ecology 50: Brassine, M. C The diet and ecological role of black-backed Jackals, Canis mesomelas, in two conservation areas in the Eastern Cape Province, South Africa. Master s thesis, University of Rhodes, South Africa. Breitenmoser, U., Breitenmoser-Würsten, C., and Capt, S Re-introduction and present status of the lynx in Switzerland. Hystrix 10: Breitenmoser, U., Ryser, A., Molinari-Jobin, A., Zimmermann, F., Haller, H., Molinari, P., and Breitenmoser-Würsten, C In: D. W. Macdonald and A. J. Loveridge, eds Biology and conservation of wild felids. New York: Oxford University Press. Breitenmoser-Würsten, C., Henschel, P. and Sogbohossou, E Caracal caracal. In: The IUCN Red List of Threatened Species [online] Available from: (Accessed 19 January 2016). Brown, J. R., Bishop, C. A., and Brooks, R. J Effectiveness of Short-Distance Translocation and its effects on Western Rattlesnakes. Journal of Wildlife Management 73: Bunnefeld, N., Linnell, J. D., Odden, J., Van Duijn, M. A. J., and Andersen, R Risk taking by Eurasian lynx (Lynx lynx) in a human dominated landscape: effects of sex and reproductive status. Journal of Zoology 270: Burns, R. J., Zemilcka, D. E., and Savarie, P. J Effectiveness of large livestock protection collars against depredating coyotes. Wildlife Society Bulletin 24: Byrne, A. W., White, P. W., McGrath, G., O Keeffe, J., and Martin, S. W Risk of tuberculosis cattle herd breakdowns in Ireland: effects of badger culling effort, density and historic largescale interventions. Veterinary research 45:

54 Chapter 1: Literature Review and Study Area Campbell, H. A., Dwyer, R. G., Wilson, H., Irwin, T. R., and Franklin, C. E Predicting the probability of large carnivore occurrence: a strategy to promote crocodile and human coexistence. Animal Conservation 18: Cardillo, M., Mace, G. M., Jones, K. E., Bielby, J., Bininda-Emonds, O. R. P., Sechrest, W., Orme, C. D. L., and Purvis, A Multiple Causes of High Extinction Risk in Large Mammal Species. Science 309: Chattha, S. A., Hussain, S. M., Javid, A., Abbas, M. N., Mahmood, S., Barq, M. G., and Hussain, M Seasonal Diet Composition of Leopard (Panthera pardus) in Machiara National Park, Azad Jammu and Kashmir, Pakistan. Pakistan Journal of Zoology 47: Chauhan, N. P. S Human casualties and livestock depredation by black and brown bears in the Indian Himalaya, Ursus 14: Conforti, V. A., and de Azevedo, F. C. C Local perceptions of jaguars (Panthera onca) and pumas (Puma concolor) in the Iguacu National Park area, south Brazil. Biological Conservation 111: Conradie, B., and Piesse, J The Effect of Predator Culling on Livestock Losses: Ceres, South Africa, Centre for Social Science Research: University of Cape Town. Constant, N A socio-ecological approach towards understanding conflict between leopards (Panthera pardus) and humans in South Africa: Implications for leopard conservation and farming livelihoods. Doctoral dissertation, Durham University, United Kingdom. Cowling, R. M., Esler, K. J., and Rundel, P. W Namaqualand, South Africa an overview of a unique winter-rainfall desert ecosystem. Plant Ecology 142: Cowling, R., and Pierce, S Namaqualand: a succulent desert. South Africa: Fernwood Press. Crawshaw, P. G Depredation of domestic animals by large cats in Brazil. Human Dimensions of Wildlife 9: Cumming, R. G Five Years' Adventures in the Far Interior of South Africa: With Notices of the Native Tribes and Savage Animals. London: John Murray. Cunningham, S. C., Gustavson, C. R., and Ballard, W. B Diet Selection of Mountain Lions in Southeastern Arizona. Journal of Range Management 3: Czech, A., and Lisle, S Understanding and Solving the Beaver (Castor fiber L.)-Human-Conflict: An Opportunity to Improve the Environment and Economy of Poland. Kataloge der Oberösterreichischen Landesmuseen 2: Dar, N. I., Minhas, R. A., Zaman, Q., and Linkie, M Predicting the patterns, perceptions and causes of human carnivore conflict in and around Machiara National Park, Pakistan. Biological Conservation 142:

55 Chapter 1: Literature Review and Study Area De Azevedo, F. C. C Food habits and livestock depredation of sympatric jaguars and pumas in the Iguacu National Park area, south Brazil. Biotropica 40: Department of Environmental Affairs National environmental management: Biodiversity Act, 2004 (Act No. 10 of 2004). Draft norms and standards for the management of damagecausing animals in South Africa. Government Gazette Desmet, P. G Namaqualand A brief overview of the physical and floristic environment. Journal of Arid Environments 70: Dickman, A. J An assessment of pastoralist attitudes and wildlife conflict Rungwa in the Rungwa-Ruaha region, Tanzania, with particular reference to large carnivores. Masters thesis, University of Oxford, United Kingdom. Dickman, A. J., Macdonald, E. W., and Macdonald, D. W A review of financial instruments to pay for predator conservation and encourage human carnivore coexistence. PNAS 108: Divyabhanusinh The end of a trail the cheetah in India. New Delhi: Banyan Books. Do Linh San, E., Malongwe, N. B., Fike, B., Somer, M. J., and Walters, M A diverse autumn diet without dominant prey for opportunistic black-backed jackals Canis mesomelas. Wildlife Biology in Practice 5: Don Carlos, A. W., Bright, A. D., Teel, T. L., and Vaske, J. J Human black bear conflict in urban areas: an integrated approach to management response. Human Dimensions of Wildlife 14: Donnelly, C. A., Woodroffe, R., Cox, D. R., Bourne, F. J., Cheeseman, C. L., Clifton-Hadley, R. S.,... and Morrison, W. I Positive and negative effects of widespread badger culling on tuberculosis in cattle. Nature 439: Driver, A., Desmet, P. G., Rouget, M., Cowling, R. M., and Maze, K. E Succulent Karoo Ecosystem Plan Biodiversity Component Technical Report. Cape Town: Cape Conservation Unit, Botanical Society of South Africa Du Plessis, J. J Towards the development of a sustainable management strategy for Canis mesomelas and Caracal caracal on rangeland. Dissertation, University of the Free State, South Africa. Du Plessis, J. J., Avenant, N. L., and De Waal, H. O Quality and quantity of the scientific information available on black-backed jackals and caracals: contributing to human predator conflict management? African Journal of Wildlife Research 45: Esler, K. J., Milton, S. J., and Dean, W. R. J Karooveld: Ecology and Management. South Africa: Briza. 32

56 Chapter 1: Literature Review and Study Area Estes, J. A., Terborgh, J., Brashares, J. S., Power, M. E., Berger, J., Bond, W. J.,... and Wardle, D. A Trophic downgrading of planet Earth. Science 333: Estes, R. D The behaviour guide to African mammals. Berkely: The University of California Press. Ferguson, J. W. H., Galpin, J. S., and De Wet, M. J Factors affecting the activity patterns of black-backed jackals Canis mesomelas. Journal of Zoology 214: Ferguson, J. W. H., Nel, J. A. J., and De Wet, M. J Social organization and movement patterns of black-backed jackals Canis mesomelas in South Africa. Journal of Zoology 199: Ferreira, N. A Sekere aspekte van die ekologie en die beheer van die rooikat (Felis caracal) in die Oranje-Vrystaat. Unpublished report, Orange Free State Provincial Adminstration, Directorate Environmental and Nature Conservation, Bloemfontein. Forbes, R. W The diet of black-backed jackal (Canis Mesomelas) on two contrasting land-use types in the Eastern Cape Province, South Africa and the validation of a new analytical method of mammalian hair identification. Maters thesis, Rhodes University, South Africa. Forsyth, D. M., Woolnough, A. P., Nimmo, D. G., Ritchie, E. G., Kennedy, M., Pople, A., and Watson, I A comment on the influence of dingoes on the Australian sheep flock. Australian Veterinary Journal 92: Frank, L. G., Woodroffe R, Ogada, M. O People and predators in the Laikipia District, Kenya. In: R. Woodroffe, S. Thirgood and A. Rabinowitz, eds People and Wildlife: Conflict or Coexistence? United Kingdom: Cambridge University Press. Gehring, T. M., Vercauteren, K. C., and Landry, J-M Livestock Protection Dogs in the 21st Century: Is an Ancient Tool Relevant to Modern Conservation Challenges? BioScience 60: Ghoddousi, A., Ghadirian, T., and Fahimi, H Status of caracal in Bahram gur Protected Area, Iran. CATnews 50: Goodrich, J. M Human tiger conflict: a review and call for comprehensive plans. Integrated Zoology 5: Graham, K., Beckerman, A. P., and Thirgood, S Human-predator-prey conflicts: ecological correlates, prey losses and patterns of management. Biological Conservation 122: Grobler, J. H Feeding behaviour of the caracal Felis caracal Schreber 1776 in the Mountain Zebra National Park. South African Journal of Zoology 16: Gurung, D. B., and Seeland, K Ecotourism in Bhutan: Extending its benefits to rural communities. Annals of Tourism research 35:

57 Chapter 1: Literature Review and Study Area Gusset, M., Swarner, M. J., Mponwane, L., Keletile, K., and McNutt, J. W Human wildlife conflict in northern Botswana: livestock predation by Endangered African wild dog Lycaon pictus and other carnivores. Oryx 43: Hayward, M. W., and Kerley, G. I Fencing for conservation: Restriction of evolutionary potential or a riposte to threatening processes? Biological Conservation 142: Hayward, M. W., Henschel, P., O Brien, J., Hofmeyr, M., Balme, G and Kerley, G. I. H Prey preferences of the leopard (Panthera pardus). Journal of Zoology 270: Henschel, P., Hunter, L., Breitenmoser, U., Purchase, N., Packer, C., Khorozyan, I., Bauer, H., Marker, L., Sogbohossou, E., and Breitenmoser-Wursten, C Panthera pardus. In: IUCN IUCN Red List of Threatened Species. Version [online] available from: (accessed 21 April 2015). Hilton-Taylor, C Red data list of Southern African plants. South Africa: National Botanical Institute. Hoare, R. E Determinants of human-elephant conflict in a land-use mosaic. Journal of Applied Ecology 36: Hoffman, M Canis mesomelas. The IUCN Red List of Threatened Species 2014: e.t3755a [online] available from: (accessed 26 October 2015). Hoffman, T. S., and O'Riain, M. J Monkey management: using spatial ecology to understand the extent and severity of human-baboon conflict in the Cape Peninsula, South Africa. Ecology and Society 17: 13. Ilemin Y., and Gürkan B Activity Patterns and Status Assessment of the Caracal in Datça and Bozburun Peninsulas, Southwestern Turkey. Zoology in the Middle East 50: Inskip, C., and Zimmerman, A Human-felid conflict: a review of patterns and priorities worldwide. Oryx 43: Jackson, R. M., and Wangchuk, R A community-based approach to mitigating livestock depredation by snow leopards. Human dimensions of wildlife 9: Jackson, R. M., Mishra, C., and McCarthy, Ale, S.B Snow leopards: conflicts and conservation. In: D. W. MacDonald and A. J. Loveridge, eds The Biology and Conservation of Wild Felids. United Kingdom: Oxford University Press. Jadhav, S., and Barua, M The Elephant Vanishes: Impact of human elephant conflict on people's wellbeing. Health & place 18: Johansson, O., McCarthy, T., Samelius, G., Andren, H., Tumursukh, L., and Mishra, C Snow leopard predation in a livestock dominated landscape in Mongolia. Biological Conservation 184:

58 Chapter 1: Literature Review and Study Area Johnson, A., Vongkhamheng, C., Hedemark, M., and Saithongdam, T Effects of humancarnivore conflict on tiger (Panthera tigris) and prey populations in Lao PDR. Animal Conservation 9: Kamler, J. F., Foght, J. L., and Collins, K Single black-backed jackal (Canis mesomelas) kills adult impala (Aepyceros melampus) African Journal of Ecology 48: Kamler, J. F., Klare, U., and Macdonald, D. W. 2012ª. Seasonal diet and prey selection of blackbacked jackals on a small-livestock farm in South Africa. African Journal of Ecology 50: Kamler, J. F., Stenkewitz, U., Klare, U., Jacobsen, N, F., and Macdonald, D. W. 2012ᵇ. Resource Partitioning Among Cape Foxes, Bat-Eared Foxes, and Black-Backed Jackals in South Africa. The Journal of Wildlife Management 76: Karanth, K. U., and Gopal, R An ecology-based policy framework for human-tiger co-existence in India. In R. Woodroffe, S. Thirgood, and A. Rabinowitz, eds People and Wildlife: Conflict or Coexistence? United Kingdom: Cambridge University Press. Kaunda, S. K. K Spatial utilization by black-backed jackals in southeastern Botswana African Zoology 36: Kaunda, S. K. K., and Skinner, J. D Black-backed jackal diet at Mokolodi Nature Reserve, Botswana. African Journal of Ecology 41: Kiffner, C., Wenner, C., La Violet, A., Yeh, K., and Kioko, J From savannah to farmland: effects of land-use on mammal communities in the Tarangire Manyara ecosystem, Tanzania. African Journal of Zoology 53: Kissui, B. M Livestock predation by lions, leopards, spotted hyenas, and their vulnerability to retaliatory killing in the Maasai steppe, Tanzania. Animal Conservation 11: Klare, U., Kamler, J. F., Stenkewitz, U., and Macdonald, D. W Diet, prey selection, and predation impact of black-backed jackals in South Africa. Journal of Wildlife Management 74: Kok, O. B Die dieetsamestelling van enkele karnivoorsoorte in die Vrystaat, Suid-Afrika. South African Journal of Science 92: Kok, O. B., and Nel, J. A. J Convergence and divergence in prey of sympatric canids and felids: opportunism or phylogenetic constraint? Biological Journal of the Linnean Society 83: Kostka, B Namaqualand - A Short History of Nearly Everything. FSM (Four Striped Mouse) Times 4:

59 Chapter 1: Literature Review and Study Area Lamarque, F., Anderson, J., Fergusson, R., Lagrange, M., Osei-Owusu, Y., and Bakker, L Human-wildlife conflict in Africa: causes, consequences and management strategies (No. 157). Food and Agriculture Organization of the United Nations (FAO). Lamprecht, J On diet, foraging behaviour and interspecific food competition of jackals in the Serengeti National Park, East Africa. Zeitschrift für Säugetierkunde 43: Letnic, M., Ritchie, E. G., and Dickman, C. R Top predators as biodiversity regulators: the dingo Canis lupus dingo as a case study Biological Reviews 87: Li, X., Buzzard, P., Chen, Y., and Jiang, X Patterns of Livestock Predation by Carnivores: Human Wildlife Conflict in Northwest Yunnan, China. Environmental Management 52: Linnell, J. D. C., Andersen, R., Andersone, Z., Balciauskas, L., Blanco, J. C., Boitani, L.,... and Wabakken, P The fear of wolves: a review of wolf attacks on people. Trondheim: NINA Oppdragsmelding 731. Linnell, J. D. C., Odden, J., Smith, M. E., Aanes, R., and Swenson, J. E Large carnivores that kill livestock: do problem individuals really exist? Wildlife Society Bulletin 27: Linnell, J. D. C., Swenson, J. E., and Andersen, R Predators and people: conservation of large carnivores is possible at high human densities if management policy is favourable. Animal Conservation 4: Loveridge, A. J., and Macdonald, D. W Niche separation in sympatric jackals (Canis mesomelas and Canis adustus). Journal of Zoology 259: Loveridge, A. J., and Nel, J. A. J Black-backed jackal. In: C. Sillero-Zubiri, M. Hoffman and D. W. Mcdonald, eds Canids: foxes, wolves, jackals and dogs status survey and conservation action plan. Switzerland and United Kingdom: IUCN/SSC Canid Specialist Group. [online]. Available: (accessed 2 December 2014). Loveridge, A. J., Wang, S. W., Frank, L. G., and Seidensticker, J People and wild felids: conservation of cats and management of conflicts In: D. W. Macdonald and A. J. Loveridge, eds Biology and conservation of wild felids. New York: Oxford University Press. MacKenzie, J. M Hunting and settlement in southern Africa. In: The empire of nature. New York: Manchester University Press. MacKinnon, J., and MacKinnon, K Review of the Protected Areas System in the Afrotropical Realm. Switzerland and United Kingdom: IUCN. 36

60 Chapter 1: Literature Review and Study Area Maclennan, S. D., Groom, R. J., Macdonald, D. W., and Frank, L. G Evaluation of a compensation scheme to bring about pastoralist tolerance of lions. Biological Conservation 142: Madden, F., and McQuinn, B Conservation s blind spot: The case for conflict transformation in wildlife conservation. Biological Conservation 178: Madhusudan, M. D Living amidst large wildlife: livestock and crop depredation by large mammals in the interior villages of Bhadra Tiger Reserve, South India. Environmental Management 31: Mann, G Aspects of the ecology of leopards (Panthera pardus) in the little Karoo, South Africa. Dissertation, Rhodes University, South Africa. Mariki, S. B., Svarstad, H., and Benjaminsen, T. A Elephants over the cliff: Explaining wildlife killings in Tanzania. Land Use Policy 44: Marker, L., and Dickman, A, J Notes on the spatial ecology of caracals (Felis caracal), with particular reference to Namibian farmlands. African Journal of Ecology 43: Marker, L., Dickman, A., and Schumann, M Using Livestock Guarding Dogs as a Conflict Resolution Strategy on Namibian Farms. Carnivore Damage Prevention News 5: Martins, Q The ecology of the leopard Panthera pardus in the Cederberg Mountains. Dissertation, University of Bristol, United Kingdom. Martins, Q., and Harris, S Movement, activity and hunting behaviour of leopards in the Cederberg mountains, South Africa. African Journal of Ecology 51: Martins, Q., Horsnell, W. G. C., Titus, W., Rautenbach, T., and Harris, S Diet determination of the Cape Mountain leopards using global positioning system location clusters and scat analysis. Journal of Zoology 283: McKenzie, A. A Co-operative hunting in the black-backed jackal Canis mesomelas. Dissertation, University of Pretoria, South Africa. Mech, D. L Is science in danger of sanctifying the wolf? Biological Conservation 150: Meriggi, A., and Lovari, S A review of wolf predation in southern Europe: does the wolf prefer wild prey to livestock? Journal of applied ecology 33: Michalski, F., Boulhosa, R., Faria, A., and Peres, C Human wildlife conflicts in a fragmented Amazonian forest landscape: determinants of large felid depredation on livestock. Animal Conservation 9: Mishra, C., Allen, P., McCarthy, T., Madhusudan, M. D., Bayarjargal, A., and Prins, H. H. T The role of incentive programs in conserving the snow leopard Uncia uncia. Conservation Biology 17:

61 Chapter 1: Literature Review and Study Area Mizutani, F Impact of leopards on a working ranch in Laikipia, Kenya. African Journal of Ecology 37: Moehlman, P. D Ecology of cooperation in canids. In: D. I. Rubenstein and R. W. Wrangham, eds Ecological Aspects of Social Evolution. New Jersey: Princeton University Press. Moolman, L. C n Vergelyking van die voedingsgewoontes van die rooikat Felis caracal binne en buite die Bergkwagga Nasionale Park. Koedoe 27: Moore, J. M., and Picker, M. D Heuweltjies (Earth Mounds) in the Clanwilliam District, Cape- Province, South- Africa 4000-Year-Old Termite Nests. Oecologia 86: Morales, J., Soria, D., Montoya, P., Pérez, B., and Salesa, M. J Caracal depereti nov. sp. y Felis aff. silvestris (Felidae, Mammalia) del Plioceno inferior de Layna (Soria, España). Estudios geológicos 59: Morehouse, A. T., and Boyce, M. S From venison to beef: seasonal changes in wolf diet composition in a livestock grazing landscape. Frontiers in Ecology and the Environment 9: Morelle, K., Lehaire, F., and Lejeune, P Spatio-temporal patterns of wildlife-vehicle collisions in a region with a high-density road network. Nature Conservation 5: Mouron, D., Désiré, G., Boisaubert, B., Lamarque, F. and Sanaa, M Recensement des collisions véhicules grands mammifères sauvages évolution entre les inventaires de et Gibier Faune Sauvage 15: Mucina, L and Rutherford, M.C. eds., The vegetation of South Africa, Lesotho and Swaziland. Pretoria: South African National Biodiversity Institute. Musiani, M. T., Muhly, T., Gates, C. C., Callaghan, C., Smith, M. E., and To.soni, E Seasonality and reoccurrence of depredation and wolf control in western North America. Wildlife Society Bulletin 33: Namaqua National Park Namaqua National Park: Park Management plan. South Africa: SANParks. Natrass, N., and Conradie, B Jackal Narratives and Predator Control in the Karoo, South Africa. Centre for Social Science Research: University of Cape Town. Nel, J. A. J., Loutit, R., and Bothma, J. Du P Prey use by black-backed jackals along a desert coast. South African Journal of Wildlife Research 27: Newsome, T. M., Dellinger, J. A., Pavey, C. R., Ripple, W. J., Shores, C. R., Wirsing, A. J., and Dickman, C. R The ecological effects of providing resource subsidies to predators. Global Ecology and Biogeography 24:

62 Chapter 1: Literature Review and Study Area Norton, P. M., and Lawson, A. B Radio tracking of leopards and caracals in the Stellenbosch area, Cape Province. South African Journal of Wildlife Research 15: Norton, P. M., Lawson, A. B., Henley, S. R., and Avery, G Prey of leopards in four mountainous areas of the south-western Cape Province. South African Journal of Wildlife Research 16: Nowell, K., and Jackson, P. eds Wild cats: status survey and conservation action plan. Gland: IUCN. Ogada, M, O., Woodroffe, R., Oguge, N. O., and Frank, L Limiting Depredation by African Carnivores: the Role of Livestock Husbandry. Conservation Biology 17: Owens, M. J., and Owens, D. D Feeding ecology and its influence on social organization in Brown hyenas (Hyaena brunnea, Thunberg) of the Central Kalahari Desert. East African Wildlife Journal 16: Palmer, R., and Fairall, N Caracal and African wild cat diet in the Karoo National Park and the implications thereof for hyrax. South African Journal of Wildlife Research 18: Parker, G. E., and Osborn, F, V Investigating the potential for chilli Capsicum spp. to reduce human-wildlife conflict in Zimbabwe. Oryx 40: Patterson, B. D., Kasiki, S. M., Selempo, E., and Kays, R. W Livestock predation by lions (Panthera leo) and other carnivores on ranches neighbouring Tsavo National Park, Kenya. Biological Conservation 119: Patterson, J. H The man-eaters of Tsavo. London: MacMillan and Company. Peterhans, J. C. K., and Gnoske, T. P The science of man-eating among lions Panthera leo with a reconstruction of the natural history of the Man-eaters of Tsavo. Journal of East African Natural History 90: Pettigrew, M., Xie, Y., Kang, A., Rao, M., Goodrich, J., Liu, T., and Berger, J Human carnivore conflict in China: a review of current approaches with recommendations for improved management. Integrative Zoology 7: Prowse, T. A. A., Johnson, C. N., Cassey, P., Bradshaw, C. J. A., and Brook, B. W Ecological and economic benefits to cattle rangelands of restoring an apex predator. Journal of Applied Ecology 52: Pyke, G. H., Pulliam, H. R., and Charnov, E. L Optimal foraging: a selective review of theory and tests. The Quarterly Review of Biology 52: Rabinowitz, A Jaguar predation on domestic livestock in Belize. Wildlife Society Bulletin 14:

63 Chapter 1: Literature Review and Study Area Rabinowitz, A Jaguars and livestock: living with the world s third largest cat. In R. Woodroffe, S. Thirgood and A. Rabinowitz, eds People and Wildlife: Conflict or Coexistence? United Kingdom: Cambridge University Press. Rajpurohit, K. S., and Krausman, P. R Human-sloth-bear conflicts in Madhya Pradesh, India. Wildlife Society Bulletin 28: Ray, J. C., Hunter, L., and Zigouris, J Setting conservation and research priorities for larger African carnivores. Wildlife Conservation Society, New York. Redpath, S. M., Gutiérrez, R. J., Wood, K. A., and Young, J. C Conflicts in Conservation: Navigating towards solutions. Cambridge: Cambridge University Press. Rowe-Rowe, D. T Food of the black-backed jackal in nature conservation and farming areas in Natal. African Journal of Ecology 11: Rowe-Rowe, D. T Black-backed jackal diet in relation to food availability in the Natal Drakensberg. South African Journal of Wildlife Research 13: Ruth, T. K., Logan, K. A., Sweanor, L. L.,Hornocker, M.G., and Temple, L. J Evaluating cougar translocation in New Mexico. Journal of Wildlife Management 62: Schiess-Meier, M., Ramsauer, S., Gabanapelo, T., and Köning, B Livestock predation insights from problem animal control registers in Botswana. Journal of Wildlife Management 71: Schumann, W., Walls, J. L., and Harley, V Attitudes towards carnivores: the views of emerging commercial farmers in Namibia. Oryx 46: Siddiqui, N. A., and Choudhury, J. H Maneating behaviour of tigers (Panthera tigris Linn) of the Sundarbans. Twenty-eight years' record analysis. Tigerpaper. Sillero-Zubiri, C., and Switzer, D Management of wild canids in human-dominated landscapes. In: Canids: foxes, wolves, jackals and dogs. Status survey and conservation action plan. United Kingdom: IUCN Canid Specialist Group, Switzerland and Cambridge. Skead, C. J Historical incidence of the larger land mammals in the broader Western and Northern Cape, (A. F. Boshoff, G. I. H. Kerley, and P. H. Lloyd, eds. Centre for African Conservation Ecology, Nelson Mandela Metropolitan University, Port Elizabeth. Skinner, J. D., and Chimimba, C. T The mammals of the southern African subregion. Cape Town: Cambridge University Press. Smith, E, R An assessment of caracal population density and human-predator conflict in the Winterberg, Eastern Cape, South Africa. Masters Thesis, Rhodes University, South Africa. Smith, M. E., Linnell, J. D., Odden, J., and Swenson, J. E Review of methods to reduce livestock depredation: I. Guardian animals. Acta Agriculturae Scandinavica 50:

64 Chapter 1: Literature Review and Study Area Smithers, R. H. N The mammals of the southern African subregion. South Africa: University of Pretoria. Stadler H Historical perspectives on the development of problem animal management in the Cape Province. In: B. Daly, W. Davies-Mostert, S. Evans, Y. Friedmann, N. King, T. Snow and H. Stadler, eds Proceedings of a workshop on holistic management of human-wildlife conflict in the agricultural sector of South Africa. Johannesburg: Endangered Wildlife Trust. Stahl, P., Vandel, J. M., Herrenschmidt, V., and Migot, P The effect of removing lynx in reducing attacks on sheep in the French Jura Mountains. Biological Conservation 101: Stander, P. E The ecology of lions and conflict with people in north-eastern Namibia. In J. V. Heerden, ed. Symposium on Lions and Leopards as Game Ranch Animals, Onderstepoort, South Africa. Namibia: Ministry of Environment and Tourism, Windhoek. Stander, P. E., Haden, P. J., Kaqece, A., and Ghau, A The ecology of asociality in Namibian leopards. Journal of Zoology 242: Stuart, C. and Wilson, V The Cats of Southern Africa. Zimbabwe: Chipingali Wildlife Trust, Bulawayo. Stuart, C. T Notes on the mammalian carnivores of the Cape Province, South Africa. Bontebok 1: Stuart, C. T Aspects of the biology of the caracal (Felis caracal Schreber 1776) in the Cape Province, South Africa. Masters thesis, University of Rhodes, South Africa. Stuart, C. T A comparison of the food of the black-backed jackal and caracal. The Naturalist 31: Stuart, C. T., and Hickman, G. C Prey of caracal Felis caracal in two areas of Cape Province, South Africa. Journal of African Ecology 105: Suryawanshi, K. R., Bhatnagar, Y. V., Redpath, S., and Mishra, C People, predators and perceptions: patterns of livestock depredation by snow leopards and wolves. Journal of Applied Ecology. 50: Swanepoel, L. H., Lindsay, P., Somers, M. J., van Hoven, W., and Dalerum, F Extent and fragmentation of suitable leopard habitat in South Africa. Animal Conservation 16: Taruvinga, A., and A. Mushunje Society's Perceptions of African Elephants and their Relative Influence towards the Conservation of Elephants. APCBEE Procedia 10: Thorn, M., Green, M., Dalerum, F., Bateman, P. W., and Scott, D. M What drives human carnivore conflict in the North West Province of South Africa? Biological Conservation 150:

65 Chapter 1: Literature Review and Study Area Thorn, M., Green, M., Scott, D., and Marnewick, K Characteristics and determinants of human-carnivore conflict in South African farmland. Biodiversity Conservation 22: Trail, J. V Quiver Trees, Phantom Orchids and Rock Splitters: The Remarkable Survival Strategies of Plants. Canada: ECW Press. Treves, A., and Karanth, K. U Human-carnivore conflict and perspectives on carnivore management worldwide. Conservation Biology 17: Treves, A., Naughton-Treves, L. I. S. A., Harper, E. K., Mladenoff, D. J., Rose, R. A., Sickley, T. A., and Wydeven, A. P Predicting human carnivore conflict: A spatial model derived from 25 years of data on wolf predation on livestock. Conservation Biology 18: Treves, A., Wallace, R. B., and White, S Participatory planning of interventions to mitigate human wildlife conflicts. Conservation Biology 23: Treves, A., Wallace, R. B., Naughton-Treves, L., and Morales, A Co-managing human wildlife conflicts: a review. Human Dimensions of Wildlife 11: Van de Ven, T. M. F. N., Tamblin, C, J., and Kerley, G. I. H Seasonal diet of black-backed jackal in the Eastern Karoo, South Africa. Journal of Arid Environments 99: Van der Merwe, I., Tambling, C. J., Thorn, M., Scott, D. M., Yarnell, R. W., Green, M., Cameron, E. Z., and Bateman, P. W An assessment of diet overlap of two mesocarnivores in the North West Province, South Africa. African Zoology 44: Van Deventer, M., and Nel, J. A. J Habitat, food, and small mammal community structure in Namaqualand. Koedoe 49: Van Niekerk, H. N The cost of predation on small livestock in South Africa by medium-sized predators. Maters thesis, University of the Free State, Bloemfontein. Van Niekerk, W., Taljaard, P. R., and De Waal, H. O Factors affecting small livestock predation in the Western Cape Province of South Africa. In AAAE and AEASA Conference, Cape Town. Van Rooyen, M. W Management of the old field vegetation in the Namaqua National Park, South Africa: conflicting demands of conservation and tourism. The Geographical Journal 168: Van Rooyen, M. W., Henstock, R., van Rooyen, N., and van der Merwe, H Plant diversity and flowering displays on old fields in the arid Namaqua National Park, South Africa. Koedoe 52: 1-7. Van Sittert, L Keeping the enemy at bay: the extermination of wild carnivora in the Cape Colony, Environmental History 3:

66 Chapter 1: Literature Review and Study Area Wagner, K, K., Schmidt, R. H., and Conover, m. R Compensation Programs for Wildlife Damage in North America. Wildlife Society Bulletin 25: Walton, L. R., and Joly, D. O Canis mesomelas. Mammalian species 715: 1-9. Wang, S. W Understanding ecological interactions among predators, ungulates and farmers in Bhutan's Jigme Singye Wangchuck National Park. Dissertation, Cornell University, USA. Wang, S. W., and Macdonald, D. W Livestock predation by carnivores in Jigme Singye Wangchuck National Park, Bhutan. Biological Conservation 129: Watkeys, M Soils of the arid south-western zone of Africa. In: W, Dean and S, Milton, eds. The Karoo. Ecological Patterns and Processes. United Kingdom: Cambridge University Press. Wilkinson, D., Smith, G. C., Delahay, R. J., and Cheeseman, C. L A model of bovine tuberculosis in the badger Meles meles: an evaluation of different vaccination strategies. Journal of Applied Ecology 41: Wilson, S. M., Madel, M. J., Mattson, D. J., Graham, J. M., Burchfield, J. A., and Belsky, J. M Natural landscape features, human-related attractants, and conflict hotspots: a spatial analysis of human-grizzly bear conflicts. Ursus 16: Woodroffe, R Predators and people: using human densities to interpret declines of large carnivores. Animal Conservation 3: Woodroffe, R., Thirgood, S., and Rabinowitz, A People and Wildlife, Conflict or Coexistence? Cambridge: Cambridge University Press. Wozencraft, W. C Order carnivora. In: D. E. Wilson and D. M. Reeder. Mammal species of the world: a taxonomic and geographic reference. Washington, DC, Smithsonian Institute Press. Yeakel, J. D., Patterson, B. D., Fox-Dobbs, K., Okumura, M. M., Cerling, T. E., Moore, J. W.,... and Dominy, N. J Cooperation and individuality among man-eating lions. Proceedings of the National Academy of Sciences 106:

67 Chapter 2: Methodology Chapter 2: Methodology 2.1. Data Collection Diet Estimation through Scat Collection and Analysis Leopard, caracal and black-backed jackal scats were collected opportunistically and along road transects from March 2014 to April 2015 (Figure 2.1). In addition to these two sampling methods caracal scats were also collected at GPS (Global Positioning System) cluster sites from radiocollared caracal (n = 8) which were visited in the field. GPS clusters were aggregations of GPS points generated when caracal spent a large amount of time within a 50 m radius. To avoid pseudoreplication only 2 scats were collected at each cluster site (Bacon, Becic, Epp and Boyce 2011). Past studies have recommended a minimum sample size of 50 scats to infer reliable results, especially for opportunistic predators (Trites and Joy 2005; Williams, Goodenough and Stafford 2012). Dietary studies mostly rely on the sampling of scats along a predetermined route as the main method of scat collection (Corbett 1989; Atkinson, Macdonald and Kamizola 2002; Glen and Dickman 2006; Do Linh San et al.2009; van der Merwe et al. 2009; Klare, Kamler, Stenkewitz and Macdonald 2010). In our study area there was a small number of roads in the Eastern section of the national park included in the study area contrasted with high number of roads on commercial farms; with the latter making road selection for transect walks difficult. In addition, female large felids generally avoid roads thereby sampling along roads only would have biased diet estimation to males (Kure 2003; Martins 2010; Palomares et al. 2012). Further, roads in the park and on some of the farms are used extensively by tourists seasonally, whereas predators are persecuted on farms, with both these aspects potentially affecting wildlife movements. Behavioural avoidance of roads and areas near roads has been documented for many predator species (Colchero et al. 2011; Rogala et al. 2011; Northrup et al. 2012). Making use of road transects as the main scat collection method would have therefore introduced potential biases and likely resulted in insufficient samples collected. Transects walks were rather used to supplement opportunistic scat collection which occurred throughout the study area on and off roads. Transect locations were selected along randomly chosen park roads and in focal camps (farm sections) that were used in a broader baseline predator ecology study in the region. 44

68 Chapter 2: Methodology Figure 2.1. Map illustrating all locations in study area where leopard (yellow), caracal (red) and black-backed jackal (blue) scat was collected. Predator scat samples were distinguished from each other by use of segmentation, size, shape and presence and size of bone shards visible (Walker 1996). Leopard scat can be identified by clear segmentation of the scats and the presence of large amounts of hair (Walker 1996). Leopard and caracal scat, like most scat deposited by felids, has clear segmentation (Walker 1996). Caracal scat is considerably smaller than leopard scat and a cut-off based on diameter was used to differentiate the two species (< 20 mm in diameter for caracal; >20 mm in diameter for leopard) [Walker 1996]. In addition, caracal scat has smaller bone shards than those often present in leopard scat. African wild cat (Felis silvestris), the other felid present in the study area, has much smaller scat which it typically buries [Walker 1996; Stuart and Stuart 2013]. Black-backed jackal scat can be identified by its size (15-20 mm in diameter) and shape (long with pointed ends) [Walker 1996; Kamler, Klare and Macdonald 2012]. Cape fox (Vulpes chama) and bat-eared fox (Otocyon megalotis) are two other canid species occurring in the study area. Care was taken to differentiate between black-backed 45

69 Chapter 2: Methodology jackal scat and fox scat. Fox scat is smaller in size and is mostly found in middens close to den sites, compared to black-backed jackal scat which is larger in size, contains more mammalian remains such as hair and bones and can mostly be found on conspicuous sites such as shrubs for marking purposes (Ferguson, Nel and De Wet 1983; Walker 1996). Black-backed jackals are social groomers and blackbacked jackal hair can often be found in analysed scats (Kaunda and Skinner 2003; Klare et al. 2010). In the laboratory, felid scat was further positively identified due to the presence of hair from the focal predator as a result of grooming (Norton, Lawson, Henley and Avery 1986; Ott, Kerley and Boshoff 2007; Martins et al.2011; Braczkowski, Watson, Coulson and Randall 2012). Because many felid species use scat as a means of territorial marking only half of each scat was collected (Martins et al. 2011). Canid species also use urine and at times the deposition of scat as a territorial marking tool (Estes 2012). Each scat collected was placed in a brown envelope with the following information: Species name, GPS location, categorical location (farm or national park), date, collection method (opportunistic, at a GPS cluster site or transect), substrate (shrub, sand, dirt) and position of scat in relation to access (middle of road, side of road, wildlife trail, no road/trail). Samples were stored with naphthalene (moth) balls and placed in a dry area until further analysis. Only a half of a scat was collected as leopards use scat deposition for territorial marking (Martins et al. 2011; Mann 2014) Caracal capture and immobilisation Eight caracal were captured, chemically immobilised and collared for this study (B. Cristescu and K. J Teichman unpublished GPS radiocollared caracal data). Research ethics approval was provided by Stellenbosch University (SU-ACUM ), University of Cape Town (2013/V30/BC), South African National Parks (CRC-2013/ ) and the Northern Cape Department of Environment and Nature Conservation (FAUNA 1157/2013 and FAUNA 1158/2013). Suitable trap locations were determined by using camera trap data, predator sign, and local knowledge from farmers. Traps were set so that they were rapidly accessible and were fitted with VHF radio-transmitters, allowing researchers to remotely monitor traps every two hours throughout the day and night, adding to the physical trap checks. Cage traps, foot snares and padded foothold traps were used for caracal capture. Cage traps have been successfully used to capture medium-sized felids, including bobcat (Knick 1990, Boitani and Powell 2012; Broman et al. 2014) and Canada lynx (Vashon et al. 2008). A combination of wire and rope-mesh single-door cage traps that were custom-built for caracal capture were used. Foot snares are considered one of the best techniques for humane felid capture and have been used for 46

70 Chapter 2: Methodology capturing various species (Mowat, Slough and Rivard 1994; Frank, Simpson and Woodroffe 2003; Balme et al.2007). Foothold traps have also been used extensively in wild felid research projects (Roelke et al. 2008, Svoboda et al.2013; Moen, Niemi, Burdett and Mech 2015). Visual, audio lures and/or bait were used to attract caracal to the traps. Captured caracals were immobilized with Zoletil (Tiletamine-Zolazepam) at 3mg/kg using a DanInject air-powered pistol. Immobilized caracals were constantly monitored by a veterinarian with regard to temperature, heart and breath rates. Caracal sex, weight and age were recorded, with the latter based on tooth wear, colouration and body size (See Appendix 2A for datasheet). The animals were fitted with GPS radio-collars (Followit, Tellus Satellite Ultra Light, Lindesberg, Sweden). These collars were chosen due to light weight (± 200g), small size and Iridium satlink option which allowed for remote transmission of data from the collar to the researchers via satellite link. This feature eliminated the need to approach the animals periodically for remote data downloads via UHF or VHF and enabled rapid field visitation of GPS clusters after the collared animal had left the site. The GPS collars utilised in this study also make use of a drop-off function which allows researchers to remotely detach collars from the predators. Collars were programmed to acquire a GPS location every three hours, 24 hours a day Diet Estimation through GPS Radio-collar Cluster Visitation Collars transmitted data remotely via every 33 h, with delays (typically <24 h) in situations when the collar failed to connect to satellites for satlink data transmission. This technology allowed prompt identification of GPS location clusters from ed location data compiled in 3-week monitoring sessions, based on a Python algorithm developed by Knopff, Knopff, Warren and Boyce (2009). Clusters were defined as 2 locations occurring in a 50-m radius within 6 days of each other. Clustered locations, where a collared animal remains for an extended period, might indicate a kill site (Knopff, Knopff, Warren and Boyce 2009; Cristescu, Stenhouse and Boyce 2015a). Because of logistical constraints including remoteness of the area, rugged terrain and number of field teams available, a subset of randomly selected clusters were visited. The geometric centroids outputted by the Python algorithm were recorded in handheld GPS-s used by field teams to navigate to the cluster sites. Each cluster site was searched systematically, on a 50 m radius commencing at the cluster centroid. Total search time was standardized to two man-hours per site, with the exception of cluster sites where shrub cover was 50%. Such situations occurred 47

71 Chapter 2: Methodology when the entire site or > 50% was ploughed field or barren land, in which case total search time for the site was reduced to one man-hour. Teams typically comprised two people, in which case the standardized total cluster search period was divided by two, with each person searching half the 50 m radius disk. The search pattern followed a zigzag, starting at the centroid and walking out to the edge of the 50 m radius (see Appendix 2B). If time was still left once the outer edge of the disk was reached, the persons zigzagged back towards the centroid, revisiting certain areas within their allocated search zone to cover these in more detail. Search teams looked for any prey remains including carcasses, bone fragments, hair, rumens, feathers and drag marks. If a prey item was located before the allotted search time was over, the location was marked in the GPS to enable revisitation and closer examination after the full cluster search was concluded. In the event that caracal scat was located during the search, a sample was collected once the search was completed. Scat samples were bagged in a brown paper envelope, labelled with the species name, GPS coordinates, categorical location (farm or national park), collection date, collection method (at a GPS cluster, including cluster ID), substrate (e.g., shrub, sand, dirt) and position of scat in relation to roads/wildlife trails (e.g., middle of trail, side of road). In the event where > 1 caracal scat was found, samples were collected from a maximum of two scats to minimize pseudo-replication (Bacon et al.2011). Scats that did not correspond to the age of the cluster (porous, old scats that easily crumbled upon applied pressure) were not collected. Photographs of prey remains were taken and the remains were used to determine the prey species (see Appendix 2C) [Skinner and Chimimba 2005, Stuart and Stuart 2007]. Hair was collected from any prey item that could not be reliably assigned to species level in the field and was later analysed under a compound microscope for species identification based on cuticle and medulla patterns (Keogh 1979; Keogh 1983; Martins et al.2011). When possible, prey sex and age were determined in the field, with age class (adult; sub-adult; YoY [Young-of-Year]) based on tooth wear (incisors and premolars) and gum recession line (Schroeder and Robb 2005). Mandible photographs were taken and later cross-referenced for age validation. When an ungulate femoral bone was located, prey body condition was assessed based on bone-marrow colour (white/yellow: good condition; pink: average condition; red: poor condition) [Yaetes, Edey and Hill 1975]. As young ungulates can have red bone-marrow due to vascularization characteristic of the bone growth process, body condition data were only collected for adult animals. 48

72 Chapter 2: Methodology Prey Abundance Estimation through Camera Trapping Camera trapping has been used in various studies to determine the relative abundance index (RAI) of a certain species or various species across a specific area (O Brien, Kinnaird and Wibisono 2003; Jenks et al.2011; Braczkowski, Watson, Coulson and Randall 2012). Camera trapping is a noninvasive research technique that allows for continuous monitoring of animal occurrence (Karanth, Nichols and Kumar 2011). For this study prey was monitored using an 810 km 2 camera trap grid which delineated the study area extent. Grid cells were 3 km 3 km squares (cell area = 9 km 2 ), with two camera trap stations used in rotation to monitor each cell. Because this study formed part of a larger project that included caracal as focal study species, cell size was selected to correspond to female caracal home range size (Avenant and Nel 1998; Martins 2010) to enable density estimation using marked (radio-collared) caracal. Each station had a single Cuddeback Ambush Black Flash camera attached to a metal post set at a standardized distance from the nearest edge of the jeep track (1 m) and at a specific lens height above the ground (0.4 m). The camera faced the jeep track perpendicularly. Initial station location was identified in a GIS (Geographic Information System) program through random generation of 2 points within each grid cell. For each point a perpendicular line to the nearest jeep track was traced in GIS, using a high resolution Google Earth image as baselayer. A point was generated at the location where the line intersected the jeep track and GPS coordinates for the point were extracted in GIS. Field teams navigated to the point and chose the final station placement by walking 100 m along the jeep track in both directions starting from the GIS-generated point, and selecting the location that maximized wildlife detection within 100 m from the initial point. Stations were preferentially placed at points where the jeep track was intersected by another jeep track, river bed, wildlife trail, or edge between distinct habitat types (e.g., shrubland and barren land/boulders, shrubland and ploughed field). All camera trap stations were set in the same broad vegetation type of Namaqualand Klipkoppe Scrubland, which is a component of the Succulent Karoo biome (Mucina and Rutherford 2006). The camera trap survey ran for 12 months from May 2014 to April A total of 89 cameras (stations) where set out for the first rotation. At the end of the rotation only 82 stations had data which we were able to use (i.e., 2 months of the camera being active). For the second rotation 87 cameras were set out and only 77 had usable data. This resulted in a total of 176 stations set, of which 159 (90.3%) had usable data. Camera failure was generally as a result of either SD card failure, battery explosion or tampering of cameras by animals, particularly baboons and cattle. 49

73 Chapter 2: Methodology Figure 2.2. Map illustrating all locations in study area camera trapping occurred each cell represents 9 km² with two camera stations for each grid cell Prey Abundance Estimation through Small Mammal Trapping Live small mammal trapping was undertaken to determine prey species available on farms and in Namaqua National Park. This was necessary because small mammals such as rodents are not reliably detected with camera traps (O Brien, Kinnaird and Wibisono 2011). Small mammal trapping occurred near camera trap stations for a period of three months from September to November Trapping only occurred in the spring as the region s semi-arid climate was the least extreme at this time of the year. Trapping was conducted at 94 of the camera trapping locations (59 farm and 35 national park) with 16 traps deployed at each location. Sherman aluminium traps (230 x 75 x 90 mm) were used and placed in a grid system with 4 rows of 4 traps each spaced 10 m from each other (see Appendix 2D). Each Sherman trap contained a piece of apple and cucumber which were kept soaked in water until placed in traps to prevent captured animals from dehydrating, a peanut butter and oats bait ball wrapped in wax paper to limit desiccation of the bait, 2 pieces of dry cat food as bait for insectivores 50

74 Chapter 2: Methodology and a small bundle of sheep wool to prevent hypothermia. Each trap was wrapped in an aerothane sheet secured to the trap with elastic bands, which provided thermal insulation during colder nights and warmer mornings. When placed in the field traps were set under shrubs to shelter the captured animals from weather. Traps were set out for a total of three trap nights per site. Traps were opened just before sunset (17h00) and checked and closed just after sunrise (07h00). Once an animal was trapped, a Ziploc transparent bag was placed over half of the trap, the door facing inside the bag was opened gently and the animal was dropped into the bag for safe handling. Captured animals were scruffed through the bag and marked by a combination of hair clipping and marking a foot with non-toxic black nail varnish. Small mammals were identified to species level using Stuart and Stuart (2007) and De Graaff (1981) and photographs were taken. Animals were sexed by distinguishing the differences in genitalia (Hoffmann et al. 2010). Each individual was also weighed in the Ziploc bag and after the release of the animal the bag was weighed again and subtracted from the original weight recorded (Hoffmann et al. 2010) Data Analysis Diet Estimation through Scat Analysis Scat samples were autoclaved at 120 C for 20 minutes to allow for complete sterilisation of samples. Autoclaved samples were individually placed in a sorting tray and sorted under a fumehood, removing macroscopic fragments (e.g., bones, insects) before washing the remains of the scat in a sieve (Cristescu, Stenhouse and Boyce 2015b). A mortar and pestle was used to help break up scats and grind faecal matter to ease the washing process. Once clean the hair was spread out on a petri dish and dried for 24 hours in the fumehood. Thereafter, hair samples were soaked in 70% ethanol for 24 hours to ensure no particles were still attached to the hairs before further analysis. Hairs were then rinsed with distilled water and dried for another 24 hours, or until dry, in the fumehood. All mammal prey categories were identified to species level by means of cross-sections of hairs. Cross-sections were made by randomly selecting hairs with a pair of forceps and placing them longitudinally in a 3 mm plastic Pasteur pipette. Forbes (2011) concluded that a minimum of 15 hairs should be used to produce a viable cross-section for analysis. Cross-sections were made by using the methods proposed by Douglas (1989). Once hairs were placed in the pipette, molten wax (Paraplast Plus, Leica Biosystems) was drawn up into the pipette after which it was immediately placed in a beaker of ice to ensure setting of the wax. Small cross-sections were then cut and mounted on glass 51

75 Chapter 2: Methodology slides using a small droplet of wax. A Leica DM 2000 light microscope was used to photograph and examine slides at 20x magnification (where possible 40x). LAS Core V4.0 software was used to measure cross-sections of the hairs for comparison with the reference collections (Rhodes University, Anita Meyer [The Cape Leopard Trust], Keogh (1979), Keogh (1983) and personal slides made from hair collected from carcasses encountered in the field). Using teeth collected from scat samples, rodents were identified to species level using de Graaff (1981) for further validation of species from cross-sections of hairs. Macroscopic and microscopic presence and absence were recorded for each scat for the following prey categories: large mammals (> 40 kg), medium- to large-sized mammals (10 40 kg), mediumsized mammals (1 10 kg), small mammals (< 1 kg) [Mann 2014], livestock, birds, reptiles, invertebrates, fruit/seeds and vegetation. In some cases an item could be recorded as unknown ungulate or unknown small mammal, however these were all grouped under unknown category to simplify results. Invertebrates were identified to order level. Reptiles were divided into lizard, tortoise and snake. Fruits/seeds, vegetation and birds were only marked as present or absent and not identified to a lower level. All mammal prey categories were identified to species level by means of cross-sections of hairs as described above. All shrews were classified as Soricidae, all hares (Lepus saxatilis, Lepus capensis) and red rock rabbits (Pronolagus rupestris) as lagomorpha, Otomys irroratus and Otomys unisulcatus were grouped into Otomys spp. and Elephantulus rupestris and Elephantulus edwardii were grouped into Elephantulus spp. The frequency of occurrence (per prey type) [FO], corrected frequency of occurrence (frequency of occurrence per scat) [CFO] and percentage biomass were calculated. FO was calculated as the number of times a prey item was recorded divided by the total number of prey items identified from all scats analysed, expressed as a percentage (Klare, Kamler and Macdonald 2011). Klare et al. (2011) recommend the use of frequency of occurrence per scat, further referred to as the CFO, where each scat has a total weighting of 1. If two prey items are present in one scat, each prey item would receive a weighting of 0.5 and less as the number of prey items per scat increases. Klare et al. (2011) concluded that the sole use of FO per prey item overestimates prey items such as invertebrates. Past studies recommend refraining from only using this method (FO) when representing diet results (Klare et al. 2011; Braczkowski et al. 2012). Various predator diet studies only present the frequency of occurrence (FO) and relative frequency of occurrence (CFO), ignoring the percentage biomass (Ott et al. 2007; Walker et al. 2007; Carrera et al. 2008; Van der Merwe et al. 2009; Braczkowski et al. 2012). Klare et al. (2011) concluded that a more in-depth representation of data is needed, especially when the diet study aims to present ecologically relevant results.et al.. Klare et al. (2011) 52

76 Chapter 2: Methodology also concluded that the best method to use when applying biomass calculation models (BCM) is linear regression based on feeding trials. If possible, they advised to use feeding trial data for the same carnivore species or a closely-related species if no other past literature is available. Biomass establishes the importance of a food item in the diet of the target animal, whereas frequency of occurrence includes rare food items (Klare et al.2011). This approach helps understand a carnivore s feeding ecology i.e.: whether it is a specialist or an opportunist. The main downfall of only using FO is that it cannot answer important ecological questions, such as the impact of predation on prey populations. However, when using BCM in human-wildlife conflict studies, it is advisable to note that these calculations could overestimate the biomass of livestock predated on (Klare et al.2011). For biomass calculations please refer to the individual focal species chapters Diet Estimation through GPS Radio-collar Cluster Visitation Eight caracal were immobilised and collared from March 2014 to April 2015 (Appendix 2E). The same method used to identify hair found in scat was used to identify hair found at cluster sites (see section 2.2.1). To ensure consistency the same prey categories that were used for scat analysis were used to group prey items identified at GPS cluster sites (also section 2.2.1). The frequency of occurrence (per prey item) [FO] and corrected frequency of occurrence (frequency of occurrence per scat) [CFO] were calculated. For a more in-depth description of FO and CFO please refer to Chapter The frequency of occurrence (per prey item) [FO] which is calculated as the number of times a prey item is recorded divided by the total number of prey items and multiplied by a 100 to calculate a percentage. Biomass was calculated by assigning an estimated weight to each prey item identified at kill sites according to age (Morehouse and Boyce 2011; Pitman et al. 2013). Where the age was marked as unknown, an average of different age weight for the prey species was estimated and used. To correct the overestimation from kill site analysis a percentage estimation of consumption was made from photographs taken at kill sites. The percentage of a prey item which was consumed differed between age and prey species. Prey weighing < 4.5 kg, such as hyrax and lagomorpha, were consumed almost entirely (90%) with the exception of the rumen, viscera and fur (Estes 2012). The corrected biomass consumed was calculated by multiplying the biomass consumed with the percentage of prey consumed. 53

77 Chapter 2: Methodology Prey Abundance and Preference Analysis Prey relative abundance was obtained from camera traps placed in the field from May 2014 to April Camera trapping is increasingly used to estimate species abundances (O Brien et al. 2003; Rautenbach 2010; Treves, Mwima, Plumptre and Isoke 2010; Jenks et al. 2011; Mann 2014). In instances where animals can be identified individually due to distinctive features or markings, markrecapture models are used. However, not all animals can be individually identified and as a result different methods are used to estimate the abundance of these animals. Due to the rugged terrain of our study area and the shy nature and low densities of animals, camera traps were the most efficient method to determine abundance. Road counts and aerial counts were not used due to the mountainous terrain and lack of reliable road networks. The relative abundance index (RAI) of each species was calculated by multiplying the total useable captures of a specific species by 100 and then dividing by the total number of trap nights (Jenks et al. 2011). RAI can also present draw-backs to data analysis as variable detection probabilities between different species are not taken into account (Sollman, Mohamed, Samejima and Wilting 2013). To avoid overestimation of animals, a time interval of 0.5 hours was used to distinguish between independent captures of a species at the same camera station (Martins 2010; Rautenbach 2010; Jenks et al. 2011). This was problematic for groupliving animals such as baboons and some ungulate species as it could underestimate the abundance (Mann 2014). Another draw-back of this method is the placement of cameras to decrease bias a randomised system was used to determine camera placement sites. Certain species such as Klipspringer, Hyrax and Red Rock Rabbit reside in rocky areas and as cameras were placed on jeep tracks these species could be underestimated. Any photographs of farm workers, hikers, vehicles or domestic dogs were excluded from the analysis. Birds were also excluded as only larger species such as bustards were captured on the cameras. Small mammal abundance was obtained from small mammal trapping with Sherman traps in the spring. Bush karoo rats (Otomys unisulcatus) are not easily trapped with Sherman traps (Cavallini and Nel 1990). Being the only method used for small mammal trapping an underestimation of Bush karoo rat abundance may have been recorded. A total of eight species of small mammals were trapped, but due to the low number of captures only data from captures of striped mouse (Rhabdomys pumillio) and Namaqua rock mouse (Aethomys namaquensis) were used and all Elephantulus spp. and round-eared sengi (Macroscelides proboscideus) were pooled as insectivores. Data were insufficient for a mark-recapture model therefore abundance was estimated as a relative abundance index (RAI), the same as for camera data. RAI was calculated as the total number of captures of a species divided by the total number of trap nights (Jenks et al. 2011). Calculations were 54

78 Chapter 2: Methodology performed separately for Namaqua National Park and the surrounding farmlands to ensure comparability between the two land-uses. Prey preference was also calculated separately for each land-use using Jacobs index (Jacobs 1974). Prey abundance data were insufficient to enable comparison of prey preference among seasons. As small mammal trapping was only conducted in spring, no seasonality data were available for small mammal abundance Statistical Analysis Diet Statistical Analysis Predator diet was analysed from 82 leopard scats, 250 caracal scats and 196 black-backed jackal scats collected in Namaqua National Park and surrounding small-stock farms. Differences in prey species and prey categories between the two land-uses in Namaqualand were tested according to a Fishers exact test (STATSoft Statistica 2008) Scat and GPS Radio-collar Cluster Visitation Statistical Analysis Diet was analysed from 250 caracal scats and 91 kill sites. Differences in scat analysis and GPS cluster visitation methods were tested using a Fishers exact test (STATSoft Statistica 2008) Prey abundance and Preference Statistical Analysis Prey abundance was calculated as the RAI (relative abundance index) for each species/prey category (e.g., lagomorpha) on both land-uses. The data were not normally distributed, so a non-parametric test was performed. Mann-Whitney-U tests (STATSoft Statistica 2008) were applied to determine whether there was a significant difference between the RAI of a certain species/prey category between Namaqua National Park and surrounding farmlands. Mann-Whitney-U tests were applied for both camera trap analysis and small mammal trapping analysis. A species accumulation curve was generated using EstimateS ver. 9 (Collwell 2013) to estimate sampling effort required for prey species detection through camera trapping and small mammal trapping. Prey preference was calculated using Jacobs index, which compares the extent to which a prey species was preyed upon to its relative availability (Jacobs 1974). The CFO was used to determine which prey species were preferred by leopards. The following equation illustrates the Jacobs index: 55

79 Chapter 2: Methodology Where is the relevant species, is the proportion of scats and is the RAI obtained from camera traps or small mammal traps. A certain prey species was preferred by the predator if 0 D 1 and avoided when -1 D 0. A D-value close to 0 would indicate prey consumption in proportion to prey availability, meaning the prey items was neither preferred, nor avoided (Jacobs 1974) References Atkinson, R. P. D., Macdonald, D. W., and Kamizola, R Dietary opportunism in side-striped jackals Canis adustus Sundevall. Journal of Zoology 257: Avenant, N. L., and Nel, J. A. J Home-range use, activity, and density of caracal in relation to prey density. African Journal of Ecology 36: Bacon, M. M., Becic, G. M., Epp, M. T., and Boyce, M. S Do GPS Clusters Really Work? Carnivore Diet from Scat Analysis and GPS Telemetry Methods. Wildlife Society Bulletin 35: Balme, G., Hunter, L., and Slotow, R Feeding habitat selection by hunting leopards Panthera pardus in a woodland savanna: prey catchability versus abundance. Animal Behaviour 74: Boitani, L., and Powell, R. A Carnivore ecology and conservation: a handbook of techniques. Oxford: Oxford University Press. Braczkowski, A., Watson, L., Coulson, D., and Randall, R Diet of leopards in the southern Cape, South Africa. African Journal of Ecology 50: Braczkowski, A., Watson, L., Coulson, D., Lucas, J., Peiser, B., and Rossi, M The diet of caracal, Caracal caracal, in two areas of the southern Cape, South Africa as determined by scat analysis. South African Journal of Wildlife Research 42: Broman, D. J. A., Litvaitis, J. A., Ellingwood, M., Tate, P., and Reed, G. C Modelling bobcat Lynx rufus habitat associations using telemetry locations and citizen-scientist observations: are the results comparable? Wildlife Biology 20: Carrera, R., Ballard, W. B., Gipson, P., Kelly, B. T., Krausman, P. R., Wallace, M. C., and Villalobos, C Comparison of Mexican wolf and coyote diets in Arizona and New Mexico. The Journal of Wildlife Management 72: Cavallini, P., and Nel, J. A. J The feeding ecology of the Cape grey mongoose, Galerella pulverulenta (Wagner 1839) in a coastal area. African Journal of Ecology 28:

80 Chapter 2: Methodology Colchero, F., Conde, D. A., Manterola, C., Chávez, C., Rivera, A., and Ceballos, G Jaguars on the move: modeling movement to mitigate fragmentation from road expansion in the Mayan Forest. Animal Conservation 14: Collwell, R. K EstimateS: Statistical estimation of species richness and shared species from samples. Version 9. Persistent URL <purl.oclc.org/estimates>. Corbett, L. K Assessing the diet of dingoes from feces: a comparison of 3 methods. The Journal of Wildlife Management 53: Cristescu, B., Stenhouse, G. B., and Boyce, M. S. 2015a. Grizzly bear diet shifting on reclaimed mines. Global Ecology and Conservation 4: Cristescu, B., Stenhouse, G. B., and Boyce, M. S. 2015b. Predicting multiple behaviors from GPS radiocollar cluster data. Behavioral Ecology 26: De Graaff, G The rodents of southern Africa. Pretoria: Buttersworth. Do Linh San, E., Malongwe, N. B., Fike, B., Somer, M. J., and Walters, M A diverse autumn diet without dominant prey for opportunistic black-backed jackals Canis mesomelas. Wildlife Biology in Practice 5: Douglas, R. M A new method of cross-sectioning hair of larger mammals. South African Journal of Wildlife Research 9: Estes, R. D The behaviour guide to African mammals. Berkely: The University of California Press. Ferguson, J. W. H., Nel, J. A. J., and De Wet, M. J Social organization and movement patterns of black-backed jackals Canis mesomelas in South Africa. Journal of Zoology 199: Forbes, R. W The diet of black-backed jackal (Canis mesomelas) on two contrasting land-use types in the Eastern Cape Province, South Africa and the validation of a new analytical method of mammalian hair identification. Masters thesis, Rhodes University, South Africa. Frank, L., Simpson, D., and Woodroffe, R. Foot Snares: An Effective Method for Capturing African Lions. Wildlife Society Bulletin 31: Glen, A. S., and Dickman, C. R Diet of the spotted-tailed quoll (Dasyurus maculatus) in eastern Australia: effects of season, sex and size. Journal of Zoology 269: Hoffmann, A., Decher, J., Rovero, F., Schaer, J., Voight, C., and Wibbelt, G Field Methods and Techniques for Monitoring Mammals In: J. Eymann, J. Degreef, Ch. Häuser, J.C. Monje, Y. Samyn and D. VandenSpiegel, eds Manual on field recording techniques and protocols for All Taxa Biodiversity Inventories and Monitoring. Belgium: ABC Taxa. Jacobs, J Quantitative measurement of food selection: a modification of the forage ratio and Ivlev s electivity index. Oecologia 14:

81 Chapter 2: Methodology Jenks, K. E., Chanteap, P., Damrongchainarong, K., Cutter, P., Cutter, P., Redford, T., Lynam, A. J., Howard, J., Leimgruber, P Using relative abundance indices from camera-trapping to test wildlife conservation hypotheses an example from Khao Yai National Park, Thailand. Tropical Conservation Science 2: Kamler, J. F., Klare, U., and Macdonald, D. W Seasonal diet and prey selection of black-backed jackals on a small-livestock farm in South Africa. African Journal of Ecology 50: Karanth, K.U., Nichols, J. D., and Kumar, N. S Estimating tiger abundance from camera trap data: field surveys and analytical issues. In: A. F. O Connell, J. D. Nichols and K. U. Karanth, eds Camera Traps in Animal Ecology: Methods and Analyses. Tokyo: Springer. Kaunda, S. K. K., and Skinner, J. D Black-backed jackal diet at Mokolodi Nature Reserve, Botswana. African Journal of Ecology 41: Keogh, H. J An atlas of hair from southern African species with reference to its taxonomic and ecological significance. Dissertation, University of Pretoria, South Africa. Keogh, H. J A photographic reference system of the microstructure of the hair of southern African bovids. South African Journal of Wildlife Research 13: Klare, U., Kamler, J. F., and Macdonald, D. W A comparison and critique of different scatanalysis methods for determining carnivore diet. Mammal Review 41: Klare, U., Kamler, J. F., Stenkewitz, U., and Macdonald, D. W Diet, prey selection, and predation impact of black-backed jackals in South Africa. Journal of Wildlife Management 74: Knick, S. T Ecology of Bobcats Relative to Exploitation and a Prey Decline in Southeastern Idaho. Wildlife Monographs 108: Knopff, K. H., Knopff, A. A., Warren, M. B., and Boyce, M. S Evaluating Global Positioning System Telemetry Techniques for Estimating Cougar Predation. The Journal of Wildlife Management 73: Kure, N Living with leopards. Cape Town: Sunbird. Mann, G Aspects of the ecology of leopards (Panthera pardus) in the little Karoo, South Africa. Dissertation, Rhodes University, South Africa. Martins, Q The ecology of the leopard Panthera pardus in the Cederberg Mountains. Dissertation, University of Bristol, United Kingdom. Martins, Q., Horsnell, W. G. C., Titus, W., Rautenbach, T., and Harris, S Diet determination of the Cape Mountain leopards using global positioning system location clusters and scat analysis. Journal of Zoology 283:

82 Chapter 2: Methodology Moen, R., Niemi, G., Burdett, C. L., and Mech, L. D Canada Lynx in the Great Lakes Region. Annual Report to USDA Forest Service and MN Cooperative Fish and Wildlife Research Unit. NRRI Technical Report No: NRRI/TR Morehouse, A. T., and Boyce, M. S From venison to beef: seasonal changes in wolf diet composition in a livestock grazing landscape. Frontiers in Ecology and the Environment 9: Mowat, G., Slough, B. G., and Rivard, R A Comparison of Three Live Capturing Devices for Lynx: Capture Efficiency and Injuries. Wildlife Society Bulletin 22: Mucina, L and Rutherford, M.C. eds., The vegetation of South Africa, Lesotho and Swaziland. Pretoria: South African National Biodiversity Institute. Northrup, J. M., Pitt, J., Muhly, T. B., Stenhouse, G. B., Musiani, M., and Boyce, M. S Vehicle traffic shapes grizzly bear behaviour on a multiple-use landscape. Journal of Applied Ecology 49: Norton, P. M., Lawson, A. B., Henley, S. R., and Avery, G Prey of leopards in four mountainous areas of the south-western Cape Province. South African Journal of Wildlife Research 16: O Brien, T. G., Kinnaird, M. F., and Wibisono, H. T Crouching tigers, hidden prey: Sumatran tiger and prey populations in a tropical forest landscape. Animal Conservation 6: O Brien, T. G., Kinniard, M. F., and Wibisono, H. T Estimation of Species Richness of Large Vertebrates Using Camera Traps: An Example from an Indonesian Rainforest In: A. F. O Connell, J. D. Nichols and K. U. Karanth, eds Camera Traps in Animal Ecology: Methods and Analyses. Tokyo: Springer Ott, T., Kerley, G. I. H., and Boshoff, A. F Preliminary observations on the diet of leopards (Panthera pardus) from a conservation area and adjacent rangelands in the Baviaanskloof region, South Africa. African Zoology 42: Palomares, F., Roques, S., Chavez, C., Silveira, L., Keller, C., Sollmann, R.,... and Jácomo, A. T. A High proportion of male faeces in jaguar populations. PloS one 7: 12. Pitman, R. T., Mulvaney, J., Ramsay, P. M., Jooste, E., and Swanepoel, L. H Global Positioning System-located kills and faecal samples: a comparison of leopard dietary estimates. Journal of Zoology 292: Rautenbach, T Assessing the diet of the Cape leopard (Panthera pardus) in the Cederberg and Gamka mountains, South Africa. Masters Thesis, Nelson Mandela Metropolitan University, South Africa. Roelke, M. E., Johnson, W.E., Millan, J., Palomares, F., Revilla, E., Rodriguez, A., Calzada, J., Ferreras, P., Leon-Vizcaino, L., Delibes, M., and O Brien, S. J Exposure to disease agents in the 59

83 Chapter 2: Methodology endangered Iberian lynx (Lynx pardinus). European Journal of Wildlife Research 54: Rogala, J. K., Hebblewhite, M., Whittington, J., White, C. A., Coleshill, J., and Musiani, M Human activity differentially redistributes large mammals in the Canadian Rockies national parks. Ecology and Society 16: 16. Online [URL] Schroeder, M. A., and Robb, L. A Criteria for gender and age. In: C, E Braun, eds Techniques for wildlife investigations and management. USA: The Wildlife Society. Skinner, J. D., and Chimimba, C. T The mammals of the southern African subregion. Cape Town: Cambridge University Press. Sollmann, R., Mohamed, A., Samejima, H., and Wilting, A Risky business or simple solution Relative abundance indices from camera-trapping. Biological Conservation 159: Stuart, C., and Stuart, M Field Guide to the Mammals of Southern Africa: Revised Edition. South Africa: Struik Publishers. Stuart, C., and Stuart, M A Field Guide to the Tracks & Signs of Southern, Central and East. Cape Town: Struik Nature. Svoboda, N. J., Belant, J. L., Beyer, D. E., Duquette, J. F., and Martin, J. A Identifying bobcat Lynx rufus kill sites using a global positioning system. Wildlife Biology 19: Treves, A., Mwima, P., Plumptre, A. J., and Isoke, S Camera-trapping forest woodland wildlife of western Uganda reveals how gregariousness biases estimates of relative abundance and distribution. Biological Conservation 143: Trites, A. W., and Joy, R Dietary analysis from fecal samples: how many scats are enough? Journal of Mammalogy 86: Van der Merwe, I., Tambling, C. J., Thorn, M., Scott, D. M., Yarnell, R. W., Green, M., Cameron, E. Z., and Bateman, P. W An assessment of diet overlap of two mesocarnivores in the North West Province, South Africa. African Zoology 44: Vashon, J. H., Meehan, A. L., Jakubas, W. J., Organ, J. F., Vashon, A. D., Mclaughlin, G. T., Matula, G. J. Jr., and Crowley, S. M Spatial Ecology of a Canada Lynx Population in Northern Main. Journal of Wildlife Management 72: Walker, C Signs of the wild: a field guide to the spoor & signs of the mammals of Southern Africa. Cape Town: Struik Publishers. Walker, R. S., Novaro, A. J., Perovic, P., Palacios, R., Donadio, E., Lucherini, M., Pia, M., López, M. S Diets of three species of Andean carnivores in high-altitude deserts of Argentina. Journal of Mammalogy 88:

84 Chapter 2: Methodology Williams, R. L., Goodenough, A. E., and Stafford, R Statistical precision of diet diversity from scat and pellet analysis. Ecological Informatics 7: Yeates, N. T. M., Edey, T. N., and Hill, M. K Animal Science: Reproduction, Climate, Meat, Wool. Australia: Pergamon Press. 61

85 Chapter 2: Methodology 2.5. Appendices Appendix 2A Datasheet filled out at each caracal capture, as provided by Dr Quinton Martins. Although the original sheet was set-up for leopard captures, the same rules apply to caracal capture. 62

86 Chapter 2: Methodology 63

87 Chapter 2: Methodology 64

88 Chapter 2: Methodology Appendix 2B GPS cluster visitation searching method on a 50 m radius. A zigzag search pattern was used, with each person starting at the centroid and walking out to the edge of the 50 m radius. 65

89 Chapter 2: Methodology Appendix 2C - Photographic example of hyrax (Procavia capensis) kill remains (left) and the data sheet filled out for each kill site visited (right). 66

90 Chapter 2: Methodology Appendix 2D Small mammal trap outline 67

91 Chapter 2: Methodology Appendix 2E Sex, weight and estimated age at time of capture for the 8 caracal radio-collared for this study. Animal IDᵃ Sex Weight(kg) Estimated ageᵇ NCM1 M 8.9 Subadult NCM2 M 12.5 Adult NCM3 M 12 Adult NCM4 M 14.4 Adult NCM5 M 10.4 Subadult NCM6 M 8 Subadult NCM7 M 15 Adult NCM8 M 10.8 Adult ᵃN = Namaqua; C = Caracal; M = Male; # = number caught ᵇ Subadult < 1.5 years and Adult > 1.5 years 68

92 Chapter 3: Leopard diet Chapter 3: The diet of leopard (Panthera pardus) in Namaqualand, South Africa 3.1. Abstract The leopard (Panthera pardus) is an apex predator and the last large carnivore still persisting in Namaqualand, Northern Cape, much of South Africa. Studies on the diet of the leopard on smallstock farms are lacking and even more so in the Northern Cape. Leopards are known to depredate on livestock and assessing the extent to which these animals predate on stock could provide important information to propose solutions to mitigate human-leopard conflict in Namaqualand. Leopards are elusive big cats, which poses challenges when studying the ecology of these animals in remote, mountainous terrains. This study used scat analysis to determine the general diet of leopards in Namaqualand, as well as compare the diet between two land-uses, namely the Namaqua National Park and surrounding small stock farmlands. Prey availability across the 810 km² study area was determined with the use of camera trapping. The data obtained from the camera traps allowed this study to compare diet with prey availability showing that leopard diet is dependent on abundant prey items. Leopards had a strong preference for hyrax (Procavia capensis) but overall obtained most biomass from livestock consumption, primarily goats and to a lower extent sheep. In Namaqua National Park hyrax and medium-sized ungulates [steenbok (Raphicerus campestris), duiker (Sylvicapra grimmia) and klipspringer (Oreotragus oreotragus)] were the main prey items in leopard diet, however on the farmlands medium-sized ungulates were replaced with livestockdue to a high level of depredation observed which could potentially lead to high persecution of leopards in Namaqualand it is crucial to implement conservation strategies to decrease livestock losses. Providing a suitable wild prey base on farmlands and increasing livestock guarding could decrease livestock losses. 69

93 Chapter 3: Leopard diet 3.2. Introduction Human-carnivore conflict is an ever increasing problem and usually arises in areas where humans and carnivores compete for the same resources and/or occupy the same area (Pettigrew et al. 2012). With the increased expansion of human development and continued urbanisation of natural areas, carnivores have been pushed out of historical ranges or forced to continue to live in close proximity to humans (Treves and Karanth 2003; Kiffner et al. 2014). Human development is encroaching on natural habitat and in many areas buffer zones between protected areas and local communities are becoming smaller (Gusset et al. 2009). In some instances buffer zones do not exist or people live inside protected areas practicing animal husbandry (Bagchi and Mishra 2006). This often results in a higher livestock biomass when compared to the natural ungulate biomass (Bagchi and Mishra 2006; Li, Buzzard, Chen and Jiang 2013). Livestock predation by carnivores is one of the main causes for human-carnivore conflict, often leading to local people engaging in retaliatory killings of carnivores. These killings have resulted in various carnivores being exterminated from certain regions; dholes (Cuon alpinus) in Bhutan (Wang and Macdonald 2006), Eurasian lynx (Lynx lynx) in Europe (Stahl, Vandel, Herrenschmidt and Migot 2001) and lions (Panthera leo) and hyenas (Crocuta crocuta) in parts of South Africa (Beinart 2003). Felids are involved in conflict with humans worldwide (Inskip & Zimmermann 2009). In particular larger felids, ranging in size from 12 kg to 235 kg, have been found to be responsible for the largest losses (Loveridge, Wang, Frank and Seidensticker, 2010). In areas where people rely on income from livestock husbandry, the loss of stock results in negative attitudes towards damage-causing animals. When subsistence farmers lose stock to carnivores the economic loss is much greater than what would be experiences by large-scale farmers (Loveridge et al. 2010). In Kenya and Zimbabwe, lions and leopards were responsible for 11-12% of annual income loss for subsistence stock farmers (Ogada, Woodroffe, Oguge and Frank 2003). Subsistence farmers lack the revenue to prevent livestock losses; in contrast various larger scale stock farmers can afford to manage predators and often lose a larger percentage of stock to causes other than carnivores, such as disease and natural disasters (Mizutani 1997; Schiess-Meier, Ramsauer, Gabanapelo and Köning 2007; Palmeira et al. 2008). Mitigating human-carnivore conflict has become a priority in conservation (Linnell et al. 1999; Can et al. 2014). Predators form an important part of ecosystems and the loss of these predators can lead to ecological perturbations (Palomares and Caro 1999; Miller et al. 2001; Beschta and Ripple 2009). To sustain larger predators, sufficient viable habitat with a suitable prey base is required (Martins 70

94 Chapter 3: Leopard diet 2010). Protected areas that fall under these criteria are scarce and as a result many large predators compete with humans for suitable habitat often resulting in conflict and lethal persecution (both legal and illegal) [Hayward et al. 2006; Martins 2010). Mammalian carnivores generally occur at low densities and have large spatial requirements making them particularly vulnerable to extirpation due to fragmentation of suitable habitat (Balme, Slotow and Hunter 2010). It is thus crucial to have effective conservation management strategies outside protected areas such as on farmlands, which occupy the most extensive land-base in Southern Africa. Mitigating losses experienced by local communities is one of the main strategies required for predator conservation. This includes improving husbandry skills by promoting herding, kraaling and guardian dogs, changing the attitudes of local people and in some cases compensation for stock losses (Johansson et al. 2015). The availability and abundance of prey, along with various other landscape attributes, play a key role in habitat selection of carnivores (Stephens and Krebs 1986; Balme, Hunter and Slotow 2007). Many mammalian carnivores are opportunistic in their feeding behaviour and will adapt to feed on the prey which is most abundant (prey abundance hypothesis) [Hopcraft, Sinclair and Packer 2005]. Alternatively, carnivores may also alter their feeding habits according to which prey items are easier to catch (landscape hypothesis) [Hopcraft et al. 2005]. A carnivore s diet can also be illustrative of the availability of prey, especially when the carnivore displays adaptable feeding behaviour (Karanth and Sunquist 1995). When analysing predator/prey relationships the latter can be used as an indicator of ecosystem functioning, along with determining the role a carnivore plays in a particular ecosystem (Klare, Kamler and Macdonald 2011; Mann 2014; Chattha et al. 2015). Dietary analyses is thus useful not only to test the extent of predation of livestock by carnivores, but also as a tool to determine which resources are required for the persistence of a certain carnivore species (Chattha et al. 2015). Leopards (Panthera pardus) are the most widespread large felids in the world. Their success across such a wide variety of habitats can be attributed to their solitary, secretive nature, their adaptability to a variety of habitats and terrain and their opportunistic feeding behaviour (Martins et al. 2011; Estes 2012). These predominantly nocturnal felids are known to have a flexible diet and mostly select prey that are widely available (Ott, Kerley and Boshoff 2007; Rautenbach 2010). A total of 92 prey items have been recorded for leopards in sub-saharan Africa (Bailey 1993; Hayward et al. 2006), ranging from invertebrates to adult eland (Taurotragus oryx) [Bailey 1993]. Leopards have high dietary needs and require between 1.6 kg to 4.9 kg of meat each day (Bothma and le Riche 1986; Stander, Haden, Kaqece and Ghau 1997; Hayward et al. 2006). High dietary requirements and adaptable foraging behaviour have earned the leopard a reputation as a livestock killer (Marker and 71

95 Chapter 3: Leopard diet Dickman 2005; Balme, Slowtow and Hunter 2009). Leopards are the smallest of the large felids (Panthera genus) and mostly select for smaller stock such as sheep, goats and calves, compared to lions and tigers which have been recorded to prey on fully grown cattle (Loveridge et al. 2010). A dramatic reduction of leopard numbers has been observed in Africa, where leopard range has been reduced by 37% (Ray et al. 2005; Balme et al. 2010). In 1986 it was estimated that only 13% of potential leopard range was within the boundaries of protected areas (MacKinnon and MacKinnon 1986; Balme et al. 2010). This further emphasizes the need to study and understand leopard ecology outside the boundaries of protected areas, going beyond the traditional approach of leopard research inside protected areas only (Balme, Lindsey, Swanepoel and Hunter 2013). Namaqualand is a semi-arid region of South Africa and many people living in the area are reliant on livestock farming as their only source of income (Allsopp, Laurent, Debeaudoin and Samuels 2007). Due to an increase in farming activities human-wildlife conflict is extensive in this region, with carnivores including leopard, caracal (Caracal caracal) and black-backed jackal (Canis mesomelas) being persecuted due to livestock losses (Stein, Fuller, Damery, Sievert and Marker 2010; Thorn, Green, Scott and Marnewick. 2013). Historically other large carnivores were present in this region, but due to eradication of these carnivores in earlier years the predator community has been altered and leopard is the only larger carnivore still persisting in this region (Skead 2011). The leopard is considered to be the apex predator in this area and as such is also a vital component of this study. Namaqualand could still provide an adequate range of habitat for large predator persistence (Loveridge and Nel 2004; Swanepoel et al.2012). Swanepoel et al. (2012) confirmed that Eastern parts of the Northern Cape, including parts of Namaqualand, are considered as suitable leopard habitat. Leopard habitat in Namaqualand is largely contiguous and conflict with farmers is the greatest threat to predator persistence in this region making conflict mitigation measures necessary. The most popular methods used by farmers to control for the loss of livestock include traps that are indiscriminate, killing non-target species such as bat-eared fox (Otocyon megalotis) and aardwolf (Proteles cristata) which do not depredate on livestock. It is estimated that up to 85% of animals caught in such traps are non-target species, or by-catch (The Cape Leopard Trust 2011). Practical and sustainable mitigation strategies could facilitate both the persistence of leopards, as well as food security for the people that live in this region. Understanding leopard diet is a first step towards the conservation of this species enabling suggestions for non-retaliatory livestock management practices in the event that leopards consume stock on Namaqualand farms. 72

96 Chapter 3: Leopard diet Aims and Objectives The main objective of this study was to provide a current account of leopard diet in Namaqualand to act as baseline data for understanding the role of leopards in this system. Diet was also compared between the Namaqua National Park and surrounding farmlands to further deepen understanding of leopard feeding ecology, especially the role of leopards in livestock predation in the area. This study hypothesises that land-use will influence prey composition in leopard diet, in addition to influencing prey categories occurring in leopard diet. This information will contribute to the compilation of a leopard management strategy for the region and assist with mitigation of conflict in the study area. Prey availability and prey preference were determined and compared between the two land-uses. Quantifying prey availability and preference will aid in understanding what effect prey availability has on diet choice of leopards in the region and whether livestock predation occurs as a response to decreased wild prey options. 3.3 Methods Study Area The study was conducted in Namaqua National Park (S E ) and the surrounding farmlands, encompassing a total area of 810 km². For a full description of the study area see Chapter 1, section Data Collection For an in-depth description of scat collection see Chapter 2, section For prey abundance estimation through camera trapping see Chapter 2, section Data Analysis Scat Analysis See Chapter 2, section for scat washing methods and methodology regarding the preparing of cross-sections and identification of mammalian hair. 73

97 Chapter 3: Leopard diet The frequency of occurrence (per prey item) [FO], corrected frequency of occurrence (frequency of occurrence per scat) [CFO] and percentage biomass were calculated. For a more in-depth description of FO and CFO refer to Chapter 2, section To estimate the biomass of prey consumed by leopards this study used Ackerman, Lindzey, and Hemker (1984) s linear regression equation to calculate a correction factor for each prey item: Where y is the weight of prey consumed per scat collected (kg/scat) and x is the average body weight of the prey item (kg) (Martins et al.2011; Mann 2014). According to Ackerman et al. (1984) prey items with an average weight of less than 2 kg cannot be corrected for digestibility as there is an assumption that such a small prey item does not comprise a whole scat; the BCM was thus only applicable for prey items weighing > 2 kg. Currently there are no feeding trial data available for leopards, however Ackerman s equation which corrects for cougar (Puma concolor) diet, with cougars being North American felids of similar size and diet range to the leopard were used (Bacon, Becic, Epp and Boyce.2011; Martins et al.2011). Prey items occurring in < 5% of a total scat was excluded from biomass calculations as these prey items usually occurred in trace amounts (Bacon et al. 2011; Mann 2014) Prey Abundance and Preference Analysis See Chapter 2, section for more information. Please note that only camera trapping analysis is applicable to this chapter Statistical Analysis For diet statistical analysis please refer to Chapter 2, section and for prey abundance and preference statistical analysis from camera trap data refer to Chapter 2, section Results Leopard diet A total of 86 leopard scats were prepared for analysis, however only 82 scats were used for analysis; 28 from Namaqua National Park and 54 from surrounding farmlands. The 4 scats that were excluded lacked hair and/or discrete bone shards. A total of 24 prey species were recorded from the 82 scats 74

98 Chapter 3: Leopard diet used in the analysis, with mammals occurring in > 90% of the total diet. According to the CFO, hyrax (Procavia capensis) [22.4%] was the prey item occurring most frequently, which together with goat (Capra hircus) [16.3%] and Lagomorpha (10.8%) made up the top three most predominant prey items (Table 3.1). Medium-sized mammals (35%), livestock (27.8%) and medium- to large mammals (21.7%) occurred most frequently in leopard diet. 75

99 Chapter 3: Leopard diet Table 3.1. Prey classes and prey species recorded in leopard scat (n=82) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Frequency of occurrence (FO) (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=100). Corrected frequency of occurrence (CFO) (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=82). For a table with all species identified see Appendix 3A. Prey Item Prey Weight (kg) Number of Occurrences (prey items) n = 100 FO (%) Number of Occurrences (per scat) n = 82 CFO (%) Large mammals (>40 kg) Medium- to large mammals (10 40 kg) Duiker (Sylvicapra grimmia) Medium mammals (1-10 kg) Hyrax (Procavia capensis) Lagomorpha Small mammals (<1 kg) Livestock Goat (Capra hircus) Sheep (Ovis aries) Birds Invertebrates Vegetation Unknown

100 Chapter 3: Leopard diet The total biomass ingested based on the 82 scats was kg, with goat (35.3%) and sheep (Ovis aries) [16.2%] making up the largest amount of biomass consumed (Table 3.2). Red hartebeest (Alcelaphus buselaphus) [14.2%], cattle (Bos taurus) [12.4%] and duiker (Sylvicapra grimmia) [6.5%] were other prey items contributing to the bulk of biomass consumed. Rock hyrax was the most frequently consumed prey item, but only made up 3.4% of total biomass consumed. Small mammals (< 1 kg) did not make up a large percentage of biomass consumed (not one species contributed > 0.10%). When converting biomass consumed to the actual biomass consumed correction factors (CFs) were used. These CFs were calculated using Ackerman s (1984) linear regression equation to help convert naïve biomass to actual biomass consumed. The total biomass consumed was kg, much lower than the naïve biomass calculated. Once the CF was applied to each prey item, goat (21.8%), hyrax (19.1%) and sheep (11.3%) were the top three prey items contributing to total biomass consumed. Lagomorpha (9.5%), duiker (8.5%) and klipspringer (Oreotragus oreotragus) [7%] rounded out the top prey items consumed in terms of actual biomass. The percentage biomass of small mammals consumed also increased from 0.2% to 1.4%. 77

101 Chapter 3: Leopard diet Stellenbosch University Table 3.2. Biomass consumed calculated from leopard scat (n=82) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 3B. Prey Item Prey Weight (kg)ᵃ Correction factor (kg/scat)ᵇ Number of Occurrences (n=93) Prey item occurrence Biomass consumed (kg)ᶜ Biomass consumed as % of all scats Total biomass consumed (kg)ᵈ Relative biomass consumed (%) Goat (Capra hircus) Hyrax (Procavia capensis) Sheep (Ovis aries) Lagomorpha Duiker (Sylivapra grimmia) Total ᵃFrom Skinner and Chimimba (2005) ᵇ From Ackerman et al. (1984), ; only for prey >2 kg ᶜPrey weight x Number of occurrences ᵈCorrection factor x Prey item occurrence 78

102 Chapter 3: Leopard diet Namaqua National Park versus surrounding farms A total of 28 scats were analysed from Namaqua National Park and 54 from surrounding farmlands. Hyrax (29.8%), duiker (16.1%), klipspringer (10.7%) and steenbok (Raphicerus campestris) [10.2%] were found to be the most frequently consumed prey items in the national park (Table 2.3). On the surrounding farmlands, goat (22.8%), hyrax (18.5%) and sheep (14.8%) were the prey items occurring most frequently in the analysed scats (Table 3.3). No invertebrates (0%) were ingested on the farmlands and no birds in the national park (0%). A significant difference was found when comparing the occurrence of goat ( = 6.72 df = 1, p = 0.028) and sheep ( = 7.13 df = 1, p = 0.046) in leopard diet across the two land-uses. In the national park medium- to large mammals (37.5%) and medium mammals (36.4%) were the prey classes found most frequently in leopard diet, however on the farms the most frequently consumed prey classes shifted to livestock (40.4%) and medium mammals (34.3%) [Figure 3.1]. Medium- to large mammals occurred significantly more in leopard diet in the national park than on the farmlands ( = 7.80, df = 1, p =0.007). Contrastingly, livestock filled the role of medium- to large mammals on farmlands and a significant difference was observed in livestock occurrence in diet when compared between the two land-uses ( = 16.44, df = 1, p = 0.000). 79

103 Chapter 3: Leopard diet Stellenbosch University Table 3.3. Prey classes and prey species recorded in leopard scat collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. FO (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences. CFO (%) was calculated as the number of occurrences per scat divided by the total number of scats collected. For a table containing a full list of species identified see Appendix 3C (Namaqua National Park) and Appendix 3D (farmlands). Namaqua National Park Farmlands Prey Item Prey Weight (kg) Number of Occurrences (prey items) n = 39 FO (%) Number of Occurrences (per scat) n = 28 CFO (%) Number of Occurrences (prey items) n = 61 FO (%) Number of Occurrences (per scat) n = 54 CFO (%) Large mammals (>40 kg) Medium- to large mammals (10 40 kg) Duiker (Sylvicapra grimmia) Klipspringer (Oreotragus oreotragus) Steenbok (Raphicerus campestris)

104 Chapter 3: Leopard diet Stellenbosch University Medium mammals (1-10 kg) Hyrax (Procavia capensis) Lagomorpha Small mammals (<1 kg) Livestock Goat (Capra hircus) Sheep (Ovis aries) Birds Invertebrates Vegetation Unknown

105 Chapter 3: Leopard diet Figure 3.1. Prey classes recorded in leopard scat (n=82) collected in Namaqua National Park (n=28) and on surrounding farmlands (n=54), Northern Cape, South Africa. CFO (%) was calculated as the number of occurrences per scat divided by the total number of scats collected. The total biomass of prey items analysed from leopard scats was kg for the national park and kg on the farms. Hyrax (30.3%), duiker (16.8%), klipspringer (12.7%) and steenbok (9.4%) made up the bulk (> 65%) of total biomass consumed in the national park (Table 3.4) while on the farmlands, goat (27.2%), sheep (20.2%), hyrax (14.2%) and lagomorpha (10.2%) made up > 70% of total biomass consumed (Table 3.5). Leopard diet on the farmlands comprised mostly livestock, particularly goat. 82

106 Chapter 3: Leopard diet Stellenbosch University Table 3.4. Biomass consumed of the five main prey items calculated from leopard scat (n=28) collected in Namaqua National Park, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 3E. Prey Item Hyrax (Procavia capensis) Duiker (Sylvicapra grimmia) Klipspringer (Oreotragus oreotragus) Steenbok (Raphicerus campestris) Red Hartebeest (Alcelaphus buselaphus) Prey Weight (kg)ᵃ Correction factor (kg/scat)ᵇ Number of Occurrences (n=34) Prey item occurrence Biomass consumed (kg)ᶜ Biomass consumed as % of all scats Total biomass consumed (kg)ᵈ Relative biomass consumed (%) Total ᵃFrom Skinner and Chimimba (2005) ᵇ From Ackerman et al. (1984), ; ; only for prey >2 kg ᶜPrey weight x Number of occurrences ᵈCorrection factor x Prey item occurrence 83

107 Chapter 3: Leopard diet Stellenbosch University Table 3.5. Biomass consumed of the five main prey items calculated from leopard scat (n=54) collected on farmlands in Namaqualand, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. For a table listing all species see Appendix 3F. Prey Item Prey Weight (kg)ᵃ Correction factor (kg/scat)ᵇ Number of Occurrences (n=59) Prey item occurrence Biomass consumed (kg)ᶜ Biomass consumed as % of all scats Total biomass consumed (kg)ᵈ Relative biomass consumed (%) Goat (Capra hircus) Sheep (Ovis aries) Hyrax (Procavia capensis) Lagomorpha Cattle (Bos taurus) Total ᵃFrom Skinner and Chimimba (2005) ᵇ From Ackerman et al. (1984), ; only for prey >2 kg ᶜPrey weight x Number of occurrences ᵈCorrection factor x Prey item occurrence 84

108 Chapter 3: Leopard diet Prey abundance and preference For national park prey abundance, data from 43 camera traps was used in the analysis. As farmlands made up a larger portion of the study area more cameras were deployed according to the grid system. Data on farmlands was collected from 120 cameras and analysed for prey abundance. Total trap nights for the study amounted to ; 5687 in the national park and on the surrounding farmlands. There was a significant difference in duiker (U = 1213, df = 1, p < 0.05), hyrax (U = , df = 1, p < 0.05) and steenbok (U = 1368, df = 1, p < 0.05) abundances when compared between the national park and the surrounding farmlands. Steenbok and duiker had a higher abundance in the national park; whereas hyrax abundance was higher on the farmlands. These prey items were some of the most frequently preyed upon species in the national park where no livestock occurred. Klipspringer, another common prey item in leopard diet inside the national park, showed no significant difference in abundance between the two land-uses (U = 2437, df = 1, p < 0.05). Lagomorpha, sheep, duiker, steenbok and porcupine (Hystrix africaeaustralis) were the main prey items in the study area across both land-uses with the highest RAI. Most of the prey items that had a RAI of < 1 were small-sized mammals such as yellow mongoose, small-spotted genet, striped polecat and meerkat. When assessing the prey preference for some of these animals a prey preference was apparent. This illustrates the potential bias that exists when calculating the D-value (Jacobs index) for prey items that occur in the diet of the study animal for < 5%. Oryx (Oryx gazella) [10.8%], steenbok (7.1%), duiker (7%) and lagomorpha (6.1%) were the four most abundant prey items in Namaqua National Park (Figure 3.2). Lagomorpha (12.5%) and duiker (6.1%) were the two most abundant wildlife prey species/groups on the farmlands. As expected there was a significant difference in livestock abundance between the two land-uses, namely sheep (U = , df = 1, p < 0.05), goat (U = , df = 1, p < 0.05) and cattle (U = , df = 1, p < 0.05). Sheep were most abundant on the farmlands (13.4%), followed by goat (4.4%) and cattle (3.7%). 85

109 Chapter 3: Leopard diet Stellenbosch University Figure 3.2. Prey relative abundance index (RAI) calculated from camera trap data collected from March 2014 April RAI was calculated as the total detections of a certain mammalian species, multiplying by 100 (to calculate the number of photo captures per 100 trap nights), and dividing by the total number of trap nights. 86

110 Chapter 3: Leopard diet Figure 3.3. Species accumulation curve (100 randomised iterations) for the entire study area (ICE Mean = 29; ACE Mean = 29), in Namaqua National Park (ICE Mean = 27.8; ACE Mean = 27.4) and the surrounding farmlands (ICE Mean = 29; ICE Mean = 29) of the 29 wild mammal prey items 1 kg in weight and livestock in the study area. From camera trapping nights, photographs of medium-to-large mammalian species were obtained which were identified to species level. Twenty nine mammals, 4 bird species and 1 reptile species (tortoise) were identified, however only mammals of medium-to-large body size were used in the analysis. A species accumulation curve, according to the number of months cameras were active, was calculated (Figure 3.3). The species accumulation curve reached an asymptote, indicating that most species were sampled (ICE mean 29). Prey preference was analysed using camera data collected from the 159 camera traps that were placed in the field for a period of 12 months (rotated once within each grid cell). When analysing prey preference for leopard diet across both land-uses, most prey items were preferred. A Jacobs index between 0.5 and 1 indicates a strong preference. Both the CFO and relative biomass consumed values were used to calculate the Jacobs index for prey preference. Mammalian prey items for which leopards displayed a strong preference (> 0.60) were hyrax, yellow mongoose (Cynictis penicillata), goat, small spotted genet (Genetta genetta), red hartebeest, striped polecat (Ictonyx striatus)[only when analysing with CFO] and klipspringer (Figure 3.4). Caracal and porcupine were the only two prey species that had a D-value of < 0 for both CFO and biomass calculations. 87

111 Chapter 3: Leopard diet Figure 3.4. Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species. A D-value close to 0 indicate prey consumption in proportion to prey availability, (prey items was neither preferred, nor avoided).the biomass consumed and the corrected frequency of occurrence (%) used to calculate the D-value are illustrated. Prey preference was also compared between the two land-uses (Table 3.6). In Namaqua National Park, goat was the prey item with the highest preference; however it was a prey item that had a RAI of 0 and an occurrence in diet of < 5%. It is suggested that prey items with similar numbers in terms of presence in the study area and in the diet, should be excluded from the analysis. However, including all prey items would allow for more accurate results for other species occurring in the diet at > 5% (Hayward et al. 2006; Mann 2014). Hyrax, klipspringer, duiker and red hartebeest were significant prey items in the leopard s diet and considered to be strongly preferred (> 0.50). Steenbok, having a very low, but positive D-value, is a prey item which was consumed in proportion to its availability and was neither preferred nor avoided. Lagomorpha s D-value only illustrated a very low avoidance (- 0.05) of the prey item, compared to caracal (-0.47) and porcupine (-0.74) which had a high avoidance value as a prey item. On the farmlands no prey item had a D-value of +1. Hyrax, goat and klipspringer were all preferred prey items with a D-value of > The D-values for small spotted genet, striped polecat and yellow mongoose emphasize the bias that exists for prey items which occur in < 5% of the diet, wherein prey items that occur only once in scats may have a D-value of +1 or close to

112 Chapter 3: Leopard diet Table 3.6. Relative abundance index (RAI) of all mammalian species recorded on the camera traps in both Namaqua National Park and the surrounding farmlands in Namaqualand, Northern Cape. The corrected frequency of occurrence (CFO) used in Jacobs Index calculations for each separate landuse is also summarised. See Appendix 3G (CFO) and Appendix 3H (biomass consumed) for comparative figure of D-values calculated for the national park and farmlands. Prey species RAI (%) in Namaqua National Park CFO (%) in Namaqua National Park Jacobs Index (D) RAI (%) on farmlands CFO (%) on farmlands Jacobs index (D) Goat Hyrax Klipspringer Meerkat Aardvark Duiker Red hartebeest Steenbok Lagomorpha Porcupine Caracal Small spotted genet Yellow mongoose Striped polecat Cattle Sheep

113 Chapter 3: Leopard diet Oryx Baboon Aardwolf Springbok Black-backed jackal African wildcat Leopard Bat-eared fox Grey mongoose Honey badger Cape fox Donkey Horse

114 Chapter 3: Leopard diet 3.5. Discussion General diet of leopards in Namaqualand In Namaqualand leopards are the apex predators and exhibit a clear opportunistic feeding behaviour. This is evident from the main prey items of leopards in Namaqualand. Hyrax (Procavia capensis), a medium-sized mammal (3.03kg) was the main prey item of leopard in Namaqualand. Hayward et al. (2006) found that worldwide leopards exhibit a preference for medium-sized ungulates (10-40 kg). However, the studies reviewed by Hayward et al. (2006) mostly focus on leopard diet in protected areas, ignoring the importance of understanding diet across various landuses (Balme et al. 2013). In other areas of South Africa leopards also prey on smaller prey items, such as hyrax, lagomorphs and even rodents (Norton et al. 1986; Rautenbach 2010; Martins et al. 2011). The selection of these smaller prey items as a food source for leopards is primarily due to these prey items being readily available (Balme et al. 2007; Rautenbach 2010). Various studies support the opportunistic feeding behaviour of leopards, where they shift their diet to select for smaller prey items when an adequate sized prey range is lacking (Bothma and Le Riche 1984; Henschel, Abernethy and White 2005; Balme et al. 2007; Braczkowski et al. 2012; Mann 2014). Hyrax, in areas where present, are a major prey source for leopards in the Western and Northern Cape (Bothma and La Riche 1994; Martins et al. 2011). The body size of leopards in my study area is unknown, but from camera data collected and one individual captured for collaring purposes it can be assumed that they might be larger than leopards in the Fynbos and Succulent Karoo biome in the Western Cape, but possibly smaller than savanna leopards in the KTP, Northern Cape (Skinner and Chimimba 2005; Martins et al. 2011; Balme, Hunter and Braczkowski 2012; Mann 2014). It is thus not surprising that in Namaqualand where hyrax is reasonably abundant it is the main wild prey item. Only one other study on leopard diet in the Northern Cape region has been undertaken. Bothma and le Riche (1994) observed leopard diet in the Augrabies Falls National Park and Kgalagadi Transfrontier Park (KTP), Northern Cape. In the Augrabies Falls National Park hyrax were one of the main prey items consumed, corresponding with other leopard diet studies throughout South Africa (Norton et al. 1986; Ott et al. 2007; Rautenbach 2010; Martins et al. 2011). In the KTP little deviation was found in terms of leopard diet compared to previous studies, but excluded the presence of rock hyrax as this prey item is not present in that region. In the Western Cape, Martins et al. (2011) showed that rock hyrax and klipspringer were the main prey items in leopard diet in the Cederberg Mountains. This also corresponded with Norton et al. s (1986) study of leopard diet using faecal 91

115 Chapter 3: Leopard diet analysis in the Clanwilliam region, Gamka Mountains, Jonkershoek region and Wemmershoek, all in the Western Cape. However, leopard presence in a region is not limited by the availability of a certain prey item, in this instance hyrax (Estes 2012). Where hyrax is not present, leopard will shift their diet to select for other available prey items, as in studies by Ott et al. (2007) and Braczkowski et al. (2012) where leopards mostly preyed on the most available medium-sized ungulate species and a larger rodent species. In Namaqualand klipspringer was the main wild prey item for preference in leopard diet, although occurring less frequently than hyrax and lagomorph in scats analysed. In the Cederberg Mountains klipspringer was also a preferred prey item for leopards (Martins et al. 2011). Leopards in Namaqualand were even found to prey on other smaller carnivores such as smallspotted genet, yellow mongoose, striped polecat and caracal. Aardvark (Orycteropus afer) was preyed on infrequently. In the present study, 24 prey items were identified in leopard diet. This compares well to studies in similar ecosystems, such as Rautenbach (2010) who identified 17 prey items, Martins et al. (2011) with 23 prey items in the Cederberg Mountains and Mann (2014) who identified 21 mammalian prey species In the Western Cape, leopards have persisted by relying on mountainous areas and natural prey, despite increased human urbanisation and agricultural practices (Martins and Martins 2006; Swanepoel et al. 2012). Only three studies analysing leopard diet in the Western Cape have observed domestic stock as a prey item. Norton et al. (1986) and Martins et al. (2011) studied leopard diet in the Cederberg Mountains, a rugged and mountainous area, where small-stock farming is practised. Mann (2014) studied leopard diet in the Little Karoo. Various previous studies have suggested that livestock predation by leopards occurs mostly opportunistically (Ott et al. 2007; Loveridge et al. 2010; Chattha et al. 2015). A large part of Namaqualand is also mountainous and these areas were included in the present study. Leopards prefer, but are not restricted, to such areas in most parts of South Africa, in particular where human practises and presence have increased (Estes 2012; Swanepoel et al. 2012). Goats (Boergoat breed) are agile climbers venturing into rugged terrain where they are presumably exposed to high risk of predation by leopards. Goat was a preferred prey item in leopard diet in Namaqualand. Preying on sheep would require a leopard to travel further from the refuge of mountains, but catching a sheep may be easier than catching a goat (Rafiq, Afzal, Jasra, Ahmad, Khan and Farooq 2010). The predominance of goat in leopard diet from scat could be attributed to a certain individual leopard specializing in goat predation, possibly as a result of overlap in leopard territory with large areas of farmlands where a high availability of goat may be present (Linnell et al. 1999; Linnell, Swenson and Anderson 2001; Loveridge et al. 2010). However, scat was collected across the entire 92

116 Chapter 3: Leopard diet study area with sampling covering the home ranges of several leopards of both sexes (B. Cristescu and K. J Teichman unpublished camera trap data). Therefore, leopards that prey on livestock might do so due to higher encounter rate with livestock than wild prey of similar size (and hence energetic reward) to livestock, ease of capturing and subduing domestic prey, or by developing a habit for catching livestock (Linnell et al. 1999; Balme et al. 2007). For future leopard studies and in particular when analysing diet to mitigate human-carnivore conflict, it would be preferable to use scat analysis in combination with GPS cluster visitation for accurate results (Martins et al. 2011; Pitman et al. 2013). GPS cluster visitation can provide more information on individual leopard diet and further the understanding of what drives leopard prey preference in Namaqualand Namaqua National Park versus surrounding farmlands Analysing diet from Namaqua National Park was expected to be more comparable to a wide range of previous leopard studies, as most studies have been done in protected areas. In the national park leopard s diet consisted mainly of hyrax and medium-sized ungulates (duiker, klipspringer and steenbok). The prey species and prey weight range coincides with Hayward et al. (2006) s main findings on leopard prey preference (10 40 kg), as well as other studies on leopard diet in South Africa (Bothma and Le Riche 1994; Ott et al. 2007; Martins et al. 2011; Braczkowski et al. 2012; Mann 2014).However, on farmlands, the percentage by which medium-sized ungulates contributed to the total biomass consumed by leopards decreased to just over 10%. Livestock however, replaced medium-sized ungulates in the diet, with small domestic stock contributing > 40% to the total biomass consumed. There are two determinants that influence the diet of an opportunistic predator. The first being which prey is the most abundant, or widely available, and the second, which prey item requires the lowest energy expenditure to prey upon (Balme et al. 2007). Livestock on farms in Namaqualand are abundant and presumably easier to capture than wild prey thereby representing an advantageous prey for leopards to tackle. With leopard diet shifting towards livestock and hyrax on farmlands, it can be suggested that these prey items are readily available on this land-use. Camera data substantiated that livestock was an abundantly available prey source with livestock abundance being higher than that of wild ungulates on farmlands. Sheep had the highest abundance across the study area, 100% of that abundance being located on farmlands. Livestock on the farmlands represent a similar abundance to medium-sized ungulates in the national park and predictably less effort is required to catch livestock. Camera data also confirmed wild ungulate abundance to be lower on the surrounding farmlands, compared to the national park. It is possible that abundance data for klipspringer and hyrax were not accurate enough to infer RAI (Relative 93

117 Chapter 3: Leopard diet Abundance Index) results confidently. Difference in detection probabilities of certain prey items could have resulted in bias when analysing the RAI (Sollman, Mohamed, Samejima and Wilting 2013). Both these animals live in rugged, rocky areas (Skinner and Chimimba 2005). With cameras being placed along jeep tracks and other linear features it is possible that camera placement would have minimized detection of some prey species, such as rock-dwelling prey. However, some cameras were placed at habitat edges between shrubland and rocky areas thereby sampling species using both habitats. Further, hyrax was observed grazing away from rocky habitat in Namaqualand; therefore cameras could have captured them outside the rocky areas which they use as safe refuge (Estes 2012). Goat hair occurred in only one scat collected in the national park and, being absent from the park, was the prey item most preferred in the park on the basis of Jacobs index. However, as leopards travel vast distances it is very probable that the goat was caught and consumed on farmlands, but the scat was deposited in the national park. Some studies have recommended excluding prey items from prey preference calculations if occurring in < 5% of the total scats, as biases can occur (Klare et al. 2010; Kamler, Klare and Macdonald, 2012). However, other studies have included all prey species in diet preference estimation (Hayward et al. 2006; Mann 2014). According to Jacobs index, only two prey items, porcupine and lagomorpha, occurring in scats collected in the national park, were avoided as prey items by leopard. Red hartebeest occurred once in scats collected in the national park and once on farmlands. With red hartebeest abundance being higher in the national park, the Jacobs was higher on farmlands than in the park. This further illustrated the draw-backs of the use of Jacobs index for rare prey items. Hyrax and klipspringer were two of the prey items that leopards showed the strongest preference for in the national park. Martins et al. (2011) found similar results in the Cederberg Mountains using both scat analysis and GPS cluster visitations. The study by Martins et al. (2011) was carried out on various land-uses, including protected areas and small private reserves. Using scat analysis as the primary method for carnivore diet estimation is useful, but with radio-collared, large carnivores GPS clusters can be identified to guide field visitation of potential kill sites (Cristescu et al. 2015b). This is an expensive and invasive technique which requires adequate sample sizes to be scientifically significant (Blame et al. 2013). Using GPS cluster visitation in combination with scat analysis has been cited as the most appropriate method for diet determination studies (Bacon et al. 2011; Cristescu, Stenhouse and Boyce 2015a). In the current study scat analysis alone yielded comparable findings on leopard diet composition to those of studies employing both scat analysis and GPS cluster visitation (Martins et al. 2011; Pitman, Swanepoel and Ramsay 2012; Tambling et al. 2012; Mann 2014; Pitman et al. 2013). 94

118 Chapter 3: Leopard diet Leopards are known for killing and sometimes feeding on smaller carnivores such as cheetah (Actinonyx jubatus), caracal, black-backed jackal (Canis mesomelas) and even genet (Stander et al. 1997; Hayward et al. 2006). The reason for killing of carnivores might be to decrease interspecific competition for resources (Loveridge et al. 2010). In Namaqualand, leopard only preyed on smaller carnivores such as small-spotted genet, striped polecat and yellow mongoose on the farmlands, with the exception of meerkat. The caracal remains analysed from one of the scats suggest that the prey was a kitten (identified by the size of claws present). The interspecific killing of other carnivores on the farmlands could suggest that competition is higher on this land-use type due to the increased available prey spectrum, resulting in an increase in carnivore numbers due to surplus prey availability. Another suggestion is that this is an opportunistic food source for leopards on farmlands in Namaqualand where natural wild prey items were lower than in the national park based on RAI results. Caracal were more abundant on the farmlands when compared to the national park, further substantiating why caracal remains were observed in leopard scat from farmlands. Leopard and caracal likely compete for food on farmland, in particular for rock hyrax which is a main prey item for both felids. Both these predators may also be drawn to farmlands due to an increased food source (livestock) and water sources (Treves et al. 2004). On the farmlands, after hyrax, lagomorpha was the main wildlife prey item which contributed a large percentage to the total biomass consumed. Opportunistic predators select for prey items which are abundant and readily available (Loveridge et al. 2010). On the farmlands, lagomorpha was the natural prey item with the largest relative abundance, higher than in the national park. Shrub and Cape hares are mixed feeders and prefer to feed on short grass (Skinner and Chimimba 2005). Hares are often seen in areas where domestic stock and wildlife regularly graze and grasses are maintained short (Skinner and Chimimba 2005). The higher abundance of lagomorphs on the farmlands could suggest why lagomorpha as a prey item occurred more frequently in leopard diet on the farmlands, than in the national park where larger-bodied prey items such as klipspringer, duiker and steenbok were readily available. Few past studies have reported lagomorphs to be an important prey species for leopards (Mitchell, Shenton and Uys 1965; Norton et al. 1986; Martins et al. 2011). In protected areas other natural prey items are readily available and prey may also have higher overall abundances, potentially resulting in lagomorphs occurring in smaller percentages in leopard diet. Mann (2014) conducted leopard research in the Gamkaberg district, Little Karoo, Western Cape. Rautenbach (2010) also ran a diet study in the Gamkaberg area and the Cederberg region. However, Mann (2014) observed more large ungulate presence in leopard diet in Gamkaberg than Rautenbach (2010). Mann (2014) argued that game farming in the area was a novel land-use and that leopard in 95

119 Chapter 3: Leopard diet this region required time to adapt to the new prey range and select for the larger ungulates as prey items. The Namaqua National park was proclaimed in 2001 and thereafter three game species were slowly introduced, namely springbok (Antidorcas marsupialis), red hartebeest and oryx. Springbok was the first new antelope species reintroduced in 2003, with red hartebeest and oryx following in In Namaqualand leopards preyed on red hartebeest, a species weighing more > 40 kg. Red hartebeest was the only introduced ungulate species occurring in leopard diet based on our samples. However, the use of scat analysis as a dietary analysis tool makes the difference between hunted and scavenged prey difficult to impossible to distinguish (Klare et al. 2010). On various accounts dead red hartebeest were found in the national park (pers. observation). Red hartebeest are waterdependant antelope, but where melons or roots are available they will utilise these resources for water (Estes 2012). Red hartebeest are also natural migrating grazers; however in Namaqualand they are restricted to the confines of the Namaqua National Park (Novellie 1990; Estes 2012). The red hartebeest remains detected in scat could have been from leopard scavenging on animal(s) that died from dehydration. The year of data collection was drought-stricken and could explain the cases of dead or sometimes dying hartebeest observed. Camera trapping is a non-invasive method which allows researchers to monitor vast areas (O Brien 2008; Swann, Kawanishi and Palmer 2010; Mann 2014). Most studies that aim to estimate abundance of carnivore prey use actual counts of prey by means of transects, spoor and scat/pellets. However, the Namaqualand study area has rugged terrain and a limited road network, making transect counts unfeasible. Camera traps were deemed the most appropriate tool to use in the study area. Using abundance data from camera traps in conjunction with scat analysis provides a noninvasive framework for determining carnivore diet and understanding the ecological role that carnivores and their prey play in the ecosystem. Previous studies suggest that the loss of an adequate prey range can cause carnivores, especially felids, to shift their diet to alternative prey (Crawshaw 2004; Loveridge et al. 2010). In South Brazil jaguars adapted their feeding behaviour to select for prey items which are more readily available (Azevedo 2008). Jaguars have been found to prefer natural prey to livestock, but when wild prey numbers were low these animals shifted their diet to more readily available prey items such as cattle (Rabinowitz and Nottingham 1986; Azvedo 2008). In areas where natural prey numbers were low, snow leopards alternatively preyed on domestic stock, resulting in 58% of snow leopard diet to be livestock (Bagchi and Mishra 2006). In Namaqualand the evidence suggests a similar pattern; leopards prefer natural prey, but in areas where an adequate natural prey range is limited and alternative prey numbers are high and unprotected leopards will shift to livestock as an abundant prey item. It is suggested that famers in the area manage their land as such to still allow natural prey for leopards to persist, in addition to 96

120 Chapter 3: Leopard diet making use of kraaling and guarding/herding methods to decrease stock losses (Johansson et al. 2015) Conclusion The results of this study confirm the opportunistic feeding behaviour exhibited by leopards. Leopards mainly preferred hyrax and medium-sized ungulates, however on the farmlands, where the availability of these prey items were in some cases lower, leopards selected for prey species which were more abundant, namely small livestock. It can thus also be concluded that the diet of leopard reflects the abundance of prey items, as well as the appropriate size range which is preferred by leopard as a prey item. Medium-sized ungulates were more abundant in the national park, where these animals were a preferred source of prey, than on the farmlands. Leopard opportunistically also fed on smaller carnivores on farmlands, potentially eliminating these animals as competitors, due to a lack of suitable prey base or due to increased encounters between carnivores on farmlands due to an influx of prey items. This study found livestock predation by leopards to be the higher than what was found in other studies on leopard diet in the Succulent Karoo biome. With leopard being the apex predator in the study area and the last remaining large carnivore in the region it is important to establish practical and long-term mitigation strategies in Namaqualand to ensure the persistence of leopard. The results from this study suggest that a suitable wild prey base can decrease depredation by leopard on farmlands. It is also important to promote cooperation from farmers and to increase farmer livelihoods by proposing solutions to decrease livestock losses References Ackerman, B. B., Lindzey, F. G., and Hemker, T. P Cougar food habits in southern Utah. The Journal of Wildlife Management 48: Allsopp, N., Laurent, C., Debeaudoin, L. M. C., and Samuels, M. U Environmental perceptions and practices of livestock keepers on the Namaqualand Commons challenge conventional rangeland management. Journal of Arid Environments 70: Bacon, M. M., Becic, G. M., Epp, M. T., and Boyce, M. S Do GPS Clusters Really Work? Carnivore Diet from Scat Analysis and GPS Telemetry Methods. Wildlife Society Bulletin 35: Bagchi, S., and Mishra, C Living with large carnivores: predation on livestock by the snow leopard (Uncia uncia). Journal of Zoology 268:

121 Chapter 3: Leopard diet Bailey, T. N The African leopard: ecology and behavior of a solitary felid. New York: Columbia University Press. Balme, G. A., Hunter, L. T. B., and Braczkowski, A Applicability of Age-Based Hunting Regulations for African Leopards. PLoS ONE 7: e Balme, G. A., Slotow, R., and Hunter, L. T Impact of conservation interventions on the dynamics and persistence of a persecuted leopard (Panthera pardus) population. Biological Conservation 142: Balme, G. A., Slotow, R., and Hunter, L. T. B Edge effects and the impact of non-protected areas in carnivore conservation: leopards in the Phinda-Mkhuze Complex, South Africa. Animal Conservation 13: Balme, G., Hunter, L., and Slotow, R Feeding habitat selection by hunting leopards Panthera pardus in a woodland savanna: prey catchability versus abundance. Animal Behaviour 74: Balme, G., Lindsey, P. A., Swanepoel, L. H., and Hunter, L. T. B Failure of research to address the rangewide conservation needs of large carnivores: leopards in South Africa as a case study. Conservation Letters 00: 1 9. Beinart, W The rise of conservation in South Africa: settlers, livestock and the environment. Oxford: Oxford University Press. Beschta, R. L., and Ripple, W. J Large predators and trophic cascades in terrestrial ecosystems of the western United States. Biological Conservation 142: Bothma, J. du P., and Le Riche, E. A. N Aspects of the ecology and behaviour of the leopard Panthera pardus in the Kalahari Desert. Koedoe 84: Bothma, J. du P., and Le Riche, E. A. N Prey preference and hunting efficiency of the Kalahari Desert leopard. In: S. D. Miller and D. D. Everett, eds. Cats of the world: Biology, conservation and management Washington, D. C: National Wildlife Federation. Bothma, J. du P., and Le Riche, E. A. N Scat analysis and aspects of defaecation in Northern Cape leopards. South African Journal of Wildlife Research 24: Braczkowski, A., Watson, L., Coulson, D., and Randall, R Diet of leopards in the southern Cape, South Africa. African Journal of Ecology 50: Can, E. O., D Cruze, D., Garschelis, D. L., Beecham, J., and Macdonald, D. W Resolving Human- Bear Conflict: A Global Survey of Countries, Experts, and Key Factors. Conservation Letters 7:

122 Chapter 3: Leopard diet Chattha, S. A., Hussain, S. M., Javid, A., Abbas, M. N., Mahmood, S., Barq, M. G., and Hussain, M Seasonal Diet Composition of Leopard (Panthera pardus) in Machiara National Park, Azad Jammu and Kashmir, Pakistan. Pakistan Journal of Zoology 47: Crawshaw, P. G Depredation of domestic animals by large cats in Brazil. Human Dimensions of Wildlife 9: Cristescu, B., Stenhouse, G. B., and Boyce, M. S. 2015a. Grizzly bear diet shifting on reclaimed mines. Global Ecology and Conservation 4: Cristescu, B., Stenhouse, G. B., and Boyce, M. S. 2015b. Predicting multiple behaviors from GPS radiocollar cluster data. Behavioral Ecology 26: De Azevedo, F. C. C Food habits and livestock depredation of sympatric jaguars and pumas in the Iguacu National Park area, south Brazil. Biotropica 40: Department of Agriculture Forestry and Fisheries Abstract of Agricultural Statistics Department of Agriculture Forestry and Fisheries [online] Available: (accessed 11 December 2014). Estes, R. D The behaviour guide to African mammals. Berkely: The University of California Press. Graupner, E. D., and Graupner, O. F Predator-prey interrelationships in a natural big game population. Final report TN.6.4.1/4F. Pretoria. In: T. N, Bailey The African leopard: ecology and behaviour of a solitary felid. New York: Columbia University Press. Grimbeek, A. M The ecology of the leopard (Panthera pardus) in the Waterberg. Masters thesis, University of Pretoria, South Africa. Gusset, M., Swarner, M. J., Mponwane, L., Keletile, K., and McNutt, J. W Human wildlife conflict in northern Botswana: livestock predation by Endangered African wild dog Lycaon pictus and other carnivores. Oryx 43: Hayward, M. W., Henschel, P., O Brien, J., Hofmeyr, M., Balme, G and Kerley, G. I. H Prey preferences of the leopard (Panthera pardus). Journal of Zoology 270: Henschel, P., Abernethy, K. A., and White, L. J. T Leopard food habits in the Lopé National Park, Gabon, Central Africa. African Journal of Ecology 43: Hopcraft, J. G. C., Sinclair, A. R. E., and Packer, C Planning for success: Serengeti lions seek prey accessibility rather than abundance. Journal of Animal Ecology 74: Inskip, C., and Zimmermann, A Human-felid conflict: a review of patterns and priorities worldwide. Oryx 43:

123 Chapter 3: Leopard diet Johansson, O., McCarthy, T., Samelius, G., Andren, H., Tumursukh, L., and Mishra, C Snow leopard predation in a livestock dominated landscape in Mongolia. Biological Conservation 184: Kamler, J. F., Klare, U., and Macdonald, D. W Seasonal diet and prey selection of black-backed jackals on a small-livestock farm in South Africa. African Journal of Ecology 50: Karanth, K. U and Sunquist, M. E Prey selection by tiger, leopard and dhole in tropical forests. Journal of Animal Ecology 64: Kiffner, C., Wenner, C., La Violet, A., Yeh, K., and Kioko, J From savannah to farmland: effects of land-use on mammal communities in the Tarangire Manyara ecosystem, Tanzania. African Journal of Zoology 53: Klare, U., Kamler, J. F., and Macdonald, D. W A comparison and critique of different scatanalysis methods for determining carnivore diet. Mammal Review 41: Klare, U., Kamler, J. F., Stenkewitz, U., and Macdonald, D. W Diet, prey selection, and predation impact of black-backed jackals in South Africa. Journal of Wildlife Management 74: Kure, N Living with leopards. Cape Town: Sunbird. Le Roux, P. G., and Skinner, J.D A note on the ecology of the leopard (Panthera pardus Linnaeus) in the Londolozi Game Reserve. African Journal of Ecology 27: Li, X., Buzzard, P., Chen, Y., and Jiang, X Patterns of Livestock Predation by Carnivores: Human Wildlife Conflict in Northwest Yunnan, China. Environmental Management 52: Linnell, D. C., Odden, J., Smith, M. E., Aanes, R., and Swenson, J. E Large Carnivores That Kill Livestock: Do "Problem Individuals" Really Exist? Wildlife Society Bulletin 27: Linnell, J. D. C., Swenson, J. E., and Andersen, R Predators and people: conservation of large carnivores is possible at high human densities if management policy is favourable. Animal Conservation 4: Loveridge A. J., and Nel J. A. J Black-backed jackal, Schreber, 1775, Least Concern (2004). In: Canids: foxes, wolves, jackals and dogs IUCN 2004 Status Survey and Conservation Action Plan 6.3. New York: IUCN. Loveridge, A. J., Wang, S. W., Frank, L. G., and Seidensticker, J People and wild felids: conservation of cats and management of conflicts In: D. W. Macdonald and A. J. Loveridge, eds Biology and conservation of wild felids. New York: Oxford University Press. MacKinnon, J., and MacKinnon, K Review of the Protected Areas System in the Afrotropical Realm. Switzerland and United Kingdom: IUCN. 100

124 Chapter 3: Leopard diet Mann, G Aspects of the ecology of leopards (Panthera pardus) in the little Karoo, South Africa. Dissertation, Rhodes University, South Africa. Marker, L., and Dickman, A, J Factors affecting leopard (Panthera pardus) spatial ecology, with particular reference to Namibian farmlands. South African Journal of Wildlife Research 35: Martins, Q and Martins, N Leopards of the Cape: conservation and conservation concerns. International Journal of Environmental studies 63: Martins, Q The ecology of the leopard Panthera pardus in the Cederberg Mountains. Dissertation, University of Bristol, United Kingdom. Martins, Q., Horsnell, W. G. C., Titus, W., Rautenbach, T., and Harris, S Diet determination of the Cape Mountain leopards using global positioning system location clusters and scat analysis. Journal of Zoology 283: Miller, B., Dugelby, B., Foreman, D., Martinez Del Rio, C., Noss, R., Phillips, M., Reading, R., Soulé, M. E., Terborgh J., and Willcox, L The importance of large carnivores to healthy ecosystems. Endangered Species Update 18: Mitchell, B. L., Shenton, J. B., and Uys, J. C. M Predation on Large Mammals in the Kafue National Park, Zambia. Zoologica Africana 1: Mizutani, F Impact of leopards on a working ranch in Laikipia, Kenya. African Journal of Ecology 37: Norton, P. M., Lawson, A. B., Henley, S. R., and Avery, G Prey of leopards in four mountainous areas of the south-western Cape Province. South African Journal of Wildlife Research 16: Novellie, P Habitat use by indigenous grazing ungulates in relation to sward structure and veld condition. Journal of the Grassland Society of Southern Africa 7: O Brien, T. G On the use of automated cameras to estimate species richness for large- and medium-sized rainforest mammals. Animal Conservation 11: Ogada, M, O., Woodroffe, R., Oguge, N. O., and Frank, L Limiting Depredation by African Carnivores: the Role of Livestock Husbandry. Conservation Biology 17: Ott, T., Kerley, G. I. H., and Boshoff, A. F Preliminary observations on the diet of leopards (Panthera pardus) from a conservation area and adjacent rangelands in the Baviaanskloof region, South Africa. African Zoology 42: Palmeira, F.B.L., Crawshaw Jr, P. G., Haddad, C. M., Ferraz, K.M.P.M.B., and Verdade, L Cattle depredation by puma (Puma concolor) and jaguar (Panthera onca) in central-western Brazil. Biological Conservation 141:

125 Chapter 3: Leopard diet Palomares, F., and Caro, T. M Interspecific killing among mammalian carnivores. American Naturalist 153: Pettigrew, M., Xie, Y., Kang, A., Rao, M., Goodrich, J., Liu, T., and Berger, J Human carnivore conflict in China: a review of current approaches with recommendations for improved management. Integrative Zoology 7: Pitman, R. T., Mulvaney, J., Ramsay, P. M., Jooste, E., and Swanepoel, L. H Global Positioning System-located kills and faecal samples: a comparison of leopard dietary estimates. Journal of Zoology 292: Pitman, R. T., Swanepoel, L. H., and Ramsay, P. M Predictive modelling of leopard predation using contextual Global Positioning System cluster analysis. Journal of Zoology 288: Rabinowitz, A., and Nottingham, B. G Ecology and behaviour of the Jaguar (Panthera onca) in Belize, Central America. Journal of Zoology 210: Rafiq, M, K., Afzal, J., Jasra, A. W., Ahmad, I., Khan, T. N., and Farooq, M. U Foraging Preferences of Free-Ranging Sheep and Goats on the Native Vegetation of Rangelands of Pubbi Hills in Pakistan. International Journal of Agriculture and Biology 12: Rautenbach, T Assessing the diet of the Cape leopard (Panthera pardus) in the Cederberg and Gamka mountains, South Africa. Masters Thesis, Nelson Mandela Metropolitan University, South Africa. Ray, J. C., Hunter, L., and Zigouris, J Setting conservation and research priorities for larger African carnivores. New York: Wildlife Conservation Society. Schiess-Meier, M., Ramsauer, S., Gabanapelo, T., and Köning, B Livestock predation insights from problem animal control registers in Botswana. Journal of Wildlife Management 71: Skead, C. J Historical incidence of the larger land mammals in the broader Western and Northern Cape. IN: A. F. Boshoff, G. I. H. Kerley, and P. H. Lloyd, eds. Port Elizabeth: Centre for African Conservation Ecology, Nelson Mandela Metropolitan University. Skinner, J. D., and Chimimba, C. T The mammals of the southern African subregion. Cape Town: Cambridge University Press. Sollmann, R., Mohamed, A., Samejima, H., and Wilting, A Risky business or simple solution Relative abundance indices from camera-trapping. Biological Conservation 159: Stahl, P., Vandel, J. M., Herrenschmidt, V., and Migot, P The effect of removing lynx in reducing attacks on sheep in the French Jura Mountains. Biological Conservation 101: Stander, P. E., Haden, P. J., Kaqece, A., and Ghau, A The ecology of asociality in Namibian leopards. Journal of Zoology 242:

126 Chapter 3: Leopard diet Stein, A. B., Fuller, T. K., Damery, D. T., Sievert, L., and Marker, L. L Farm management and economic analyses of leopard conservation in north-central Namibia. Animal Conservation 13: Stephens, D. W and Krebs, J. R Foraging Theory. New Jersey: Princeton University Press. Swanepoel, L. H., Lindsay, P., Somers, M. J., van Hoven, W., and Dalerum, F Extent and fragmentation of suitable leopard habitat in South Africa. Animal Conservation 16: Swann, D. E., Kawanishi, K., and Palmer, J Evaluating Types and Features of Camera Traps in Ecological Studies: A Guide for Researchers. In: A. F. O Connell, J. D. Nichols and K. U. Karanth, eds Camera Traps in Animal Ecology: Methods and Analyses. Tokyo: Springer. The Cape Leopard Trust (CLT) The Cape leopard trust annual report Cape Town: The Cape Leopard Trust. Thorn, M., Green, M., Scott, D., and Marnewick, K Characteristics and determinants of human-carnivore conflict in South African farmland. Biodiversity Conservation 22: Treves, A., and Karanth, K. U Human-carnivore conflict and perspectives on carnivore management worldwide. Conservation Biology 17: Treves, A., Naughton Treves, L. I. S. A., Harper, E. K., Mladenoff, D. J., Rose, R. A., Sickley, T. A., and Wydeven, A. P Predicting human carnivore conflict: A spatial model derived from 25 years of data on wolf predation on livestock. Conservation Biology 18: Wang, S. W., and Macdonald, D. W Livestock predation by carnivores in Jigme Singye Wangchuck National Park, Bhutan. Biological Conservation 129:

127 Chapter 3: Leopard diet 3.8. Appendices Appendix 3A - Prey items recorded in leopard scat (n=82) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Frequency of occurrence (FO) (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=100). Corrected frequency of occurrence (CFO) (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=82). Prey Item Prey Weight (kg) Number of Occurrences (prey items) n = FO (%) Number of Occurrences (per scat) n = 82 Large mammals (>40 kg) Red Hartebeest (Alcelaphus buselaphus) Aardvark (Orycteropus afer) Medium- to large mammals (10 40 kg) Duiker (Sylvicapra grimmia) Caracal (Caracal caracal) Klipspringer (Oreotragus oreotragus) Steenbok (Raphicerus campestris) Medium mammals (1-10 kg) Porcupine (Hystrix africaeaustralis) Hyrax (Procavia capensis) Lagomorpha Small spotted genet (Genetta genetta) Small mammals (<1 kg) Yellow Mongoose (Cynictis penicillata) Striped polecat (Ictonyx striatus) Meerkat (Suricate suricate) Otomys spp Hairy-footed gerbil (Gerbillurus paeba) Soricidae Livestock Cattle (Bos taurus) CFO (%)

128 Chapter 3: Leopard diet Goat (Capra hircus) Sheep (Ovis aries) Birds Invertebrates Coleoptera Scorpiones Vegetation Unknown

129 Chapter 3: Leopard diet Stellenbosch University Appendix 3B - Biomass consumed calculated from leopard scat (n=82) collected in Namaqua National Park and surrounding farmlands, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Prey Item Prey Weight (kg)ᵃ Correction factor (kg/scat)ᵇ Number of Occurrences (n=93) Prey items occurrence Biomass consumed (kg)ᶜ Biomass consumed as % of all scats Total biomass consumed (kg)ᵈ Relative biomass consumed (%) Goat (Capra hircus) Hyrax (Procavia capensis) Sheep (Ovis aries) Lagomorpha Duiker (Sylivapra grimmia) Klipspringer (Oreotragus oreotragus) Red Hartebeest (Alcelaphus buselaphus) Cattle (Bos taurus) Steenbok (Raphicerus campestris) Birds Aardvark (Orycteropus afer) Caracal (Caracal caracal) Porcupine (Hystrix africaeaustralis)

130 Chapter 3: Leopard diet Stellenbosch University Small spotted genet (Genetta genetta) Yellow Mongoose (Cynictis penicillata) Striped Polecat (Ictonyx striatus) Meerkat (Suricate suricate) Otomys spp Hairy-footed gerbil (Gerbillurus paeba) Soricidae Total ᵃFrom Skinner and Chimimba (2005) ᵇ From Ackerman et al. (1984), ; only for prey >2 kg ᶜPrey weight x Number of occurrences ᵈCorrection factor x Prey items occurrence 107

131 Chapter 3: Leopard diet Appendix 3C - Prey items recorded in leopard scat collected in Namaqua National Park, Northern Cape, South Africa. Frequency of occurrence (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=39). Corrected frequency of occurrence (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=28). Prey Item Prey Weight (kg) Number of Occurrences (prey items) n = 39 FO (%) Number of Occurrences (per scat) n = 28 Large mammals (>40 kg) Red Hartebeest (Alcelaphus buselaphus) Aardvark (Orycteropus afer) Medium- to large mammals (10 40 kg) Duiker (Sylvicapra grimmia) Caracal (Caracal caracal) Klipspringer (Oreotragus oreotragus) Steenbok (Raphicerus campestris) Medium mammals (1-10 kg) Porcupine (Hystrix africaeaustralis) Hyrax (Procavia capensis) Lagomorpha Small spotted genet (Genetta genetta) Small mammals (<1 kg) Yellow Mongoose (Cynictis penicillata) Striped polecat (Ictonyx striatus) Meerkat (Suricate suricate) Otomys spp Hairy-footed gerbil (Gerbillurus paeba) Soricidae Livestock Cattle (Bos taurus) Goat (Capra hircus) Sheep (Ovis aries) Birds Invertebrates Coleoptera Scorpiones Vegetation Unknown CFO (%) 108

132 Chapter 3: Leopard diet Appendix 3D - Prey items recorded in leopard scat collected on farmlands in Namaqualand, Northern Cape, South Africa. Frequency of occurrence (%) was calculated as the number of occurrences of each prey item divided by the total number of occurrences (n=61). Corrected frequency of occurrence (%) was calculated as the number of occurrences per scat divided by the total number of scats collected (n=54). Prey Item Prey Weight (kg) Number of Occurrences (prey items) n = 61 FO (%) Number of Occurrences (per scat) n = 54 Large mammals (>40 kg) Red Hartebeest (Alcelaphus buselaphus) Aardvark (Orycteropus afer) Medium- to large mammals (10 40 kg) Duiker (Sylvicapra grimmia) Caracal (Caracal caracal) Klipspringer (Oreotragus oreotragus) Steenbok (Raphicerus campestris) Medium mammals (1-10 kg) Porcupine (Hystrix africaeaustralis) Hyrax (Procavia capensis) Lagomorpha Small spotted genet (Genetta genetta) Small mammals (<1 kg) Yellow Mongoose (Cynictis penicillata) Striped polecat (Ictonyx striatus) Meerkat (Suricate suricate) Otomys spp Hairy-footed gerbil (Gerbillurus CFO (%) paeba) Soricidae Livestock Cattle (Bos taurus) Goat (Capra hircus) Sheep (Ovis aries) Birds Invertebrates Coleoptera Scorpiones Vegetation Unknown

133 Chapter 3: Leopard diet Stellenbosch University Appendix 3E - Biomass consumed calculated from leopard scat (n=28) collected in Namaqua National Park, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Prey Item Prey Weight (kg)ᵃ Correction factor (kg/scat)ᵇ Number of Occurrences (n=34) Prey items occurrence 110 Biomass consumed (kg)ᶜ Biomass consumed as % of all scats Total biomass consumed (kg)ᵈ Relative biomass consumed (%) Hyrax (Procavia capensis) Duiker (Sylvicapra grimmia) Klipspringer (Oreotragus oreotragus) Steenbok (Raphicerus campestris) Red Hartebeest (Alcelaphus buselaphus) Lagomorpha Goat (Capra hircus) Aardvark (Orycteropus afer) Porcupine (Hystrix africaeaustralis) Suricate (Suricate suricate) Otomys spp Soricidae Yellow Mongoose (Cynictis penicillata) Hairy-footed gerbil (Gerbillurus paeba)

134 Chapter 3: Leopard diet Stellenbosch University Sheep (Ovis aries) Cattle (Bos taurus) Caracal (Caracal caracal) Striped Polecat (Ictonyx striatus) Small spotted genet (Genetta genetta) Birds Total ᵃFrom Skinner and Chimimba (2005) ᵇ From Ackerman et al. (1984), ; only for prey >2 kg ᶜPrey weight x Number of occurrences ᵈCorrection factor x Prey items occurrence 111

135 Chapter 3: Leopard diet Stellenbosch University Appendix 3F - Biomass consumed calculated from leopard scat (n=54) collected on farmlands in Namaqualand, Northern Cape, South Africa. Both the biomass consumed and the total biomass consumed is presented. Prey Item Prey Weight (kg)ᵃ Correction factor (kg/scat)ᵇ Number of Occurrences (n=59) Prey items occurrence Biomass consumed (kg)ᶜ Biomass consumed as % of all scats Total biomass consumed (kg)ᵈ Relative biomass consumed (%) Goat (Capra hircus) Sheep (Ovis aries) Hyrax (Procavia capensis) Lagomorpha Cattle (Bos taurus) Duiker (Sylvicapra grimmia) Klipspringer (Oreotragus oreotragus) Red Hartebeest (Alcelaphus buselaphus) Steenbok (Raphicerus campestris) Birds Caracal (Caracal caracal) Small spotted genet (Genetta genetta) Yellow Mongoose (Cynictis penicillata) Striped Polecat (Ictonyx striatus) Hairy-footed gerbil

136 Chapter 3: Leopard diet Stellenbosch University (Gerbillurus paeba) Aardvark (Orycteropus afer) Suricate (Suricate suricate) Porcupine (Hystrix africaeaustralis) Soricidae Otomys spp Total ᵃFrom Skinner and Chimimba (2005) ᵇ From Ackerman et al. (1984), ; only for prey >2 kg ᶜPrey weight x Number of occurrences ᵈCorrection factor x Prey items occurrence 113

137 Chapter 3: Leopard diet Appendix 3G - Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species in Namaqua National Park and the surrounding farmlands. D-values are based on corrected frequency of occurrence (%) of prey items from leopard scat. 114

138 Chapter 3: Leopard diet Appendix 3H - Jacobs Index (D-value) showing preference (+ 1) and avoidance (- 1) for prey species in Namaqua National Park and the surrounding farmlands. D-values are based on the total biomass consumed of prey items from leopard scat. 115